Salesforce Admins Podcast

Today on the Salesforce Admins Podcast, we talk to Jennifer Lee, Joshua Birk, and Kate Lessard from the Admin Evangelist team at Salesforce. Join us as we revisit the team’s predictions from the beginning of the year for how Agentforce will change the game for admins in 2025.  You should subscribe for the full episode, but here are a few takeaways from our conversation with Jennifer Lee, Joshua Birk, and Kate Lessard. Agentforce content highlights from 2024 I started by asking the team which content from 2024 still holds up. Between “Automate This” and “How I Solved It”, Jenn puts a lot of great stuff out there. However, she pointed to her modular flows walkthrough on the blog. By breaking complicated processes down into smaller chunks, you make it easier to reuse bits and pieces of them in future solutions. Kate was more focused on the big picture. In her blog, “Introduction to Agentforce for Salesforce Admins,” she explains why admins are the perfect candidates to become the go-to AI expert in their organizations. Unsurprisingly, Josh got a little more technical with his answer, highlighting the growing importance of Retrieval Augmented Generation (RAG) and simple prompt engineering. He points to his interview with Nochum Klein about how he uses Agentforce to organize and search information security documentation at Salesforce. How Agentforce will help admins in 2025 The team also looked ahead to 2025, and I think it’s fun to look back on how things are shaping up now that it’s July. Kate was focused on how Agentforce will affect admins’ core responsibilities. The agents you build make life easier for both you and your users. However, she pointed out that security and AI governance will be critical as it becomes easier for more people to interact with your data. Jen was excited to launch two new video series in 2025. If you haven’t yet checked out “Automate with Agentforce”, it’s been incredibly helpful in showing all the cool new solutions you can build with AI. She also has a series about how she’s learning Agentforce, which is a great place to get started. Finally, Josh was excited about building AI agents that interact with documentation and metadata, enabling faster support, onboarding, and troubleshooting experiences. When combined with Slack integration, you can save your users so much time. What we’ll be saying at the end of 2025 To finish out the episode, I asked the team to make predictions for what we’ll be saying at the end of 2025. Josh: “It’s the end of 2025, and I can’t believe Salesforce Admins found Agentforce so easy to work with.” Kate: “It’s the end of 2025 and I can’t believe Salesforce Admins are creating dynamic experiences this advanced!” Jen: “It’s the end of 2025 and I can’t believe Salesforce Admins can now do things like troubleshoot user management issues faster than ever before!” It’s halfway through the year now, so how did we do? And how is your 2025 going? Are you working with Agentforce? Navigating new AI tools? Hit us up in the Trailblazer Community and share your admin wins and lessons. Podcast swag Salesforce Admins on the Trailhead Store Learn more Jen’s 2024 blog post: Embrace Modular Flows to Build Smarter Automation for Agentforce Kate’s 2024 blog post: Introduction to Agentforce for Salesforce Admins Kate’s other 2024 blog post: Advance Your Admin Career With Dev Fundamentals Josh’s 2024 Salesforce Admins Podcast Episode: How Agentforce Transforms Customer Interactions at Salesforce Blog: 6 Tips To Help You Troubleshoot Agentforce With Confidence Blog: How Admins Drive Innovation With Core Responsibilities in the Agentforce Era Video Series: Automate with Agentforce Video series: Automate This! Video series: How I Solved It Admin Trailblazers Group Admin Trailblazers Community Group Social Jen on LinkedIn Josh on LinkedIn Kate on LinkedIn Salesforce Admins on LinkedIn Salesforce Admins on X Mike on Bluesky social Mike on Threads Mike on Tiktok Mike on X Love our podcasts? Subscribe today or review us on iTunes!   Full show transcript   Mike Gerholdt: Hey, Salesforce Admins! It’s July, and we’re officially halfway through 2025—which makes it the perfect time to hit pause and reflect. Back on January 1, we gathered the Admin Evangelist Team—Jen Lee, Kate Lassard, and Josh Burke—for a special kickoff episode full of predictions, priorities, and plans for the year ahead. So today, we’re rebroadcasting that conversation as a mid-year check-in. Were we right? Were we way off? You decide. Give it a listen, and then let us know how your year as a Salesforce Admin is shaping up. Stay tuned—and see how far we’ve come. Welcome, everybody, to the podcast. There's a lot of people to introduce, so I'm going to go in reverse order. Jen, let's start off with you. Can you give us a brief introduction and some of the cool content you've created last year at Salesforce?   Jennifer Lee: Sure, absolutely. I am Jen Lee, Lead Admin Evangelist, and you all probably know me as the host of Automate This or How I Solved It on our Salesforce Admins YouTube channel or for reading my mega blog for each release. Some of the things I'm really excited that I created last year was, again, I love Automate This. I love the How I Solved It, bringing in trailblazers who showcase their skills and what they've built in their orgs, and I really enjoyed writing the blog on building modular flows and thinking about really chunking out and building out smaller flows to get ready for your company in moving over to Agentforce.   Mike Gerholdt: Nice and got Agentforce in the first minute of the show. Kate, you're our newest member. Let's go with you next.   Kate Lassard: Hi, everyone. Kate Lassard, also a Lead Admin Evangelist here at Salesforce, and I have been here since August, so still diving in. This past year, I've been really focused on Agentforce and have put together some content including an intro to Agentforce blog post, talking about why admins make great AI specialists, and then also talking about advancing your admin career with dev fundamentals. So you might've seen me on the road at one of the Agentforce tours talking about core responsibilities, and in the new year, keep your eyes open for some new content about how emerging AI technologies fit into those admin core responsibilities.   Mike Gerholdt: New stuff in the new year. I like it. And of course, Josh Burke.   Josh Birk: Hi, everybody. I think actually I'm technically the oldest member of the team, but that's only by chronological age. I've been on the admin team for, gosh, I think it's a little over a year now, but I've been at Salesforce for coming on 15 years in 2025. All of them in evangelism in one form or another. And a lot of the things I've been trying to write about and post about and blog and video and some of our podcasts is really trying to explain through some of the more technical side of artificial intelligence. We have all of these terms. We've got things like LLMs, we've got RAGs, we've got vector databases, and honestly, frequently the concepts are far more simple than the tech terms actually seem to suggest. I'm on record for saying I don't like the term prompt engineering, for instance, because it sounds like you need some kind of union guy to come over and rewire your computer in order to, but is basically just talking to a conversational UI in the first place. So definitely see more of that in the near future, especially as our Agentforce features keep expanding into things like RAG and being able to pull in your knowledge libraries and your documentations and actually have a conversation with them.   Mike Gerholdt: Awesome. Well, and of course anytime somebody goes to Trailhead, Josh, they're using something you invented.   Josh Birk: I get a penny every time.   Mike Gerholdt: Oh, a penny.   Josh Birk: A whole penny.   Mike Gerholdt: A whole penny. That's before taxes. Okay. Well even in 2025. So speaking of which, we would be like, if you're listening to this the day it comes out, you're in the second day of paying for a gym membership that you think you're going to use for the rest of the month, bet you're probably not going to make it to the 15th, most of us anyway. Or eating vegetables. That's usually the two things. At least those were my horribly tried out New Year's resolutions. But I'm going to start off with Jen. So Jen, it's 2025. We've got 363 days ahead of us as Salesforce admins. What are you, as of now, going to start focusing on?   Jennifer Lee: Well, of course, Agentforce. Who isn't? I am very excited for two video series that I'm working on and hoping to put out soon, and this will focus around one will focus around automation and thinking ahead about how that factors into Agentforce and the strategy that you should think about and work through in bringing that along with your company. So that's going to be, I think about 10 to 12 episodes. We're looking to do monthly and then we'll be going through, and just like you all are learning Agentforce, I'm learning Agentforce, so I'm going to take you on my journey of how I'm working through various pieces. So really excited to put that type of video content out.   Mike Gerholdt: Yeah, we're excited to see it. Kate, similar question, but a little bit different. You were most recently in the seat as an admin. What are some things you should be thinking of as a Salesforce admin in January that would help set you up for success for the rest of the year?   Kate Lassard: That is a great question. With all the advancements in AI and with, as I said, admins being the ideal candidates at their organizations to become their internal AI specialists due to their unique understanding of business and user needs combined with their declarative Salesforce skillset, I'm going to be paying attention to new ways of managing admin responsibilities we already have in place. So one that is top of mind for me is security. New technology like Agentforce brings advanced value to admins, but it also brings new security concerns and the need for AI governance, so I have my eyes on the innovative ways that our admin community will continue to evolve in their roles while they're addressing these emerging technologies.   Mike Gerholdt: Yeah, that's good. Josh, you probably work on some of the most advanced stuff, and half of the words that you said in your previous answer I didn't understand. Are there things that you're looking ahead at 2025 to understand that you weren't on your roadmap for 2024?   Josh Birk: Yeah. Well, not to repeat myself too much, but it's been an interesting journey with our friend RAG, which is a Retrieval Augmented Generation, which is a very fancy way of saying that the AI models can absorb information that was not part of their original training, their original model. And typically this is going to be in the form of things like PDFs and documents and things like that. And the reason why I think it's an interesting journey is when we first started talking about RAG here internally at Salesforce, it was actually more about how our models were going to start getting trained on enterprise data. So we have all of this wonderful custom metadata, and it tells the models of what your custom objects look like, what your custom fields look like, and so they can consume that using RAG in a very nice and flexible way, and you don't have to rebuild an entire LLM for it. We didn't talk about it much back then because that was behind the scenes, under the covers. This is how the engine is running kind of thing. Now it's actually turning into a very common use case where people are putting in 500 page documents. Think about that mega blog that Jen was talking about. Think about having release notes available to you through a conversational UI kind of thing, so it's something that's fastly growing. And when it comes to those like, oh, what's that killer use case that we could get in to have Agentforce really do good things for our company, RAG is turning in one of those big solutions.   Mike Gerholdt: And it's another acronym for us to learn.   Josh Birk: Right.   Mike Gerholdt: Let's pivot. We do a lot of things as events at Salesforce, and I know our admins, I always try to make it to a lot of events, but you can't make it to them all, right? FOMO is a thing. As you're planning your year, and Kate, I'll start off with you just to mix up the questions a little bit. How would you, as a Salesforce admin for 2025, look at events and what you could or could not go to? And Katie, bar the door. There's no restrictions.   Kate Lassard: Oh my goodness. Well, I would love to go to all of them. FOMO is real, but not a possibility. I think it's really about prioritizing what you want to learn. So at the beginning of every year, I always try to think about what are my learning goals? What do I want to come out of this year? And whether it's a specific certification or something like learning more about Agentforce and AI governance and security, to go back to my last answer. And then finding the events that are in alignment with that. So obviously things like TDX and Dreamforce are going to be great options because there's going to be a huge amount of content, but the community conferences are also fantastic, and figuring out which ones thematically match your learning goals is really the way that I like to approach events and always maybe trying to fit in a fun new destination like Irish Dreaming this year. So if you've never been to Ireland, maybe that's a good one to add to your event bucket list.   Mike Gerholdt: Wow, budget. Budget. Katie, bar.   Kate Lassard: I didn't say my event bucket list.   Mike Gerholdt: Yeah. No, that's cool. I mean, I love all of those Dreaming events. And Jen, to pivot to you, you speak at a lot of the Dreaming events. What are some of the things that, as an admin, you looked for and got out of going to some of these community-run conferences?   Jennifer Lee: It's a different vibe from attending Salesforce events like TDX and Dreamforce, it's more low-key, but you're actually learning from the practitioner. So it's beyond the things that Salesforce is focusing on, but getting those really best practices and things like that from the people on the ground who are doing the thing that you're doing. And that's what really excites me about going to community events because you're able to learn from your peers and talk to them and ask questions. There might be something that you're working on that you ran into roadblock, but then you attend a session and then that opened your eyes and gave you ideas and inspired you to go back and try different things, so that's why I love going to community events and just seeing all the people in the community.   Mike Gerholdt: And the time in between sessions is usually the most fun because it's when you connect with everybody.   Jennifer Lee: Exactly.   Mike Gerholdt: I love that. Josh, I'm not going to ask you to pick between your two favorite children of TDX and Salesforce, but I know you've been a part of both and building activations for both. I think very real in the minds of Salesforce admins is how do I justify going to one or both? What are some of the considerations that you would give admins as advice for planning their travel to one or both of those this year?   Josh Birk: Yeah. Well, I think it's important to point out that we've kind of acknowledged TDX's role as more of a builder-centric conference, something that's really about enablement, and it's about knowledge, and it's about learning, and it's really about upscaling your career and your skill set. And so I think that's one justification if you're trying to convince your boss that you really need to go to both is that one's a really good learning experience, and the other one is a really good networking experience. Not that you're not going to learn from Dreamforce, not that you're not going to get the good sessions and the good breakouts and all of that, but it is definitely, we are kind of trying to make TDX a little bit more of its own thing on the map as opposed to just kind of a companion event to Dreamforce itself. And the advice I always give people is prepare, prepare, prepare. It's just like go to agenda builder, make sure you know which sessions ahead of time that you're really going to get the most bang for your buck out of. And always that constant reminder, if breakouts don't have repeats, you might want to show up early because if that's the session that you convinced your boss to send you on the plane for, make sure that you get a seat. So yeah, no, very much looking forward to them this year. Also looking forward to the community events. I'll echo what Jen said. So I used to joke, I'm not really a developer, I just play one on TV. I guess I'm not really a developer, but I just play one on TV. But it's like we need to hear from you. You're the people on the front lines. You're the people who are actually putting these use cases together. You're going to be the people finding the weird little niche things about these features that maybe when we kick the tires of them, we didn't consider it. So it's a great way to get that wonderful feedback loop kind of closed in.   Mike Gerholdt: Yeah, no, I hear you. Plus going to TDX means sometimes you can eat at the restaurants around Moscone Center, right? There's some really good ones there. Always food. This podcast always has food, food and time travel, which is what we're going to do. So last question, and this is for everybody, and now everybody's sweating like, "Oh God, don't call on me first." Okay, Mike takes an hour and a half to ask a question. You have plenty of time. You can read War and Peace in the time it takes for me to ask a question, but we often time travel on the podcast. So we're going to fast-forward. It's now the end of December 2025. You go back and listen to this podcast, and Josh, I'm going to start with you because I didn't start with you on any of the questions. You have to complete this sentence. It's the end of 2025, and I can't believe Salesforce admins blank.   Josh Birk: Can't believe Salesforce admins found Agentforce so easy to work with. And I can kind of say that safely because it's something I've seen on the road a lot, and it's part of our job is to make it like when you say, "I'm going to go develop an artificial intelligence custom agent." It sounds like something that you better put on your scholarly hat and really dig in deep. What we're finding is it's just really not that hard. So what I'm hoping is that as we do these enablement workshops and as we get the Trailhead Playgrounds, and people can go in and kick the tires, and they just want to give that a shout-out that that's here in the present, not just in the future, that you can go get a free version of this, and you can go to Trailhead, and you can start learning these things now. And I remember back when Lightning One components hit, and everybody's like, "Oh, what do we do?" It's like, "What do we do about LWC?" It's like, well, don't panic, but now is the time to learn it. Now is the time. And one of the things I've said many times in my keynotes is like, now is the time to determine your relationship with AI. It's your time to figure out what's going to make you more efficient, what's going to make you more productive, what's going to make your job happier.   Mike Gerholdt: Okay, well, that was a great answer. Kate, you have to follow Frank Sinatra.   Kate Lassard: Oh my gosh.   Mike Gerholdt: So it's the end of 2025, and I can't believe Salesforce admins.   Kate Lassard: Are creating dynamic user experiences this advanced. I think that a few years ago when Salesforce announced dynamic forms, that was such a game changer for admins allowing us to create more customized dynamic user experiences right on those record pages for our users. And with Agentforce and Prompt Builder, that adds completely new functionality that admins can leverage to really create those dynamic experiences for their users, for their customers. And I think we're going to see not only a resurgence of creating those dynamic user experiences, but now we have even more capability to do so, so I can't wait to see how advanced and how exciting those experiences are.   Mike Gerholdt: Yeah, a hackathon at the end of 2025 is probably going to look a lot different than the hackathon we did in New York back in November.   Kate Lassard: Exactly.   Mike Gerholdt: Yeah. Jen, you get the final word. It's the end of 2025, and I can't believe Salesforce admins.   Jennifer Lee: Right. So not to copy what Josh or Kate said.   Mike Gerholdt: Oh, it's the hardest when you're the third person on a panel. Ditto. That's what you should say. Ditto. And the first guy just went ahead and went right to AI.   Josh Birk: I know.   Jennifer Lee: I would say, I can't believe Salesforce admins can now do things like troubleshoot user management issues faster than ever before. Just knowing what Cheryl's team is doing behind the scenes and the things that they're working on, your mind's going to be blown. We're going to have agents that help you troubleshoot those things so that you are not spending all the time trying to figure out why Josh has this permission, but Kate doesn't.   Kate Lassard: Great answer. Way to close with a bang.   Mike Gerholdt: You know what's going to be fun is I'm going to save this, and we're going to do this again at the end of 2025 and see how close we were. Oh, because why not? And then hopefully we'll say, "Yeah, so we could do that by June." Maybe not. Who knows? Thanks all for coming on the podcast and helping admins get ready for the new year. I know we have a lot of content lined up, so I appreciate it, and I appreciate your perspective on bringing things.   Josh Birk: Thanks for having us. You bet.   Jennifer Lee: Happy New Year. 2025 here we come.   Mike Gerholdt: What a ride, right? It’s always fascinating to look back and see what held true—and what surprised us. Huge thanks again to Jen, Kate, and Josh for setting the tone back in January. And now, we want to hear from you: how’s your 2025 going? Are you working with Agentforce? Navigating new AI tools? Hit us up in the Trailblazer Community and share your admin wins and lessons. Until next time, we’ll see you in the cloud.  

Today on the Salesforce Admins Podcast, we talk to Brinkal Janani, Director of Product Management at Salesforce. Join us as we chat about how Generative Canvas will help admins create dynamic, personalized user experiences. You should subscribe for the full episode, but here are a few takeaways from our conversation with Brinkal Janani. What is Generative Canvas? Brinkal was on the team of engineers that built Dynamic Forms. Since then, he’s been looking for new ways to help admins build dynamic user experiences without code. That’s why I was so excited to sit down with him and hear about his latest project: Generative Canvas. Generative Canvas allows Salesforce Admins to create interactive, persistent layouts by prompting AI agents. Basically, you ask an agent to run an analysis or summarize some records and it’ll respond with a Lightning Component that you can drop into your layout. Admins configure the agents, connect the data, and suddenly, their users can build their own dynamic, personalized UX. Persistent and personalized user experiences One of the biggest challenges for admins is anticipating what your users are going to need in terms of data and workflows. Static tools like Lightning Page Builder and Dynamic Forms are great as long as you have the right requirements. But making adjustments means a lot of back-and-forth, especially when you have to balance the needs of several different users. Generative Canvas UIs are persistent, but they’re also personalized. The admin still has control over what data can be used, but the user has control over how they see it. Instead of going through all those extra steps, they just need to ask an agent for what they want and drop the Lightning Component directly into their own individual, personalized UI. A hybrid future for admins Brinkal envisions a hybrid future where static and dynamic tools coexist. Admins might start building with Lightning pages, but move into Generative Canvas when deeper interaction is required. This hybrid approach ensures flexibility while harnessing the power of AI-driven customization. If all of this sounds a little vague to you, I highly recommend watching the demo video to understand what it looks like in action. As Brinkal says, the future is dynamic, personalized, and built with no code. For more about Generative Canvas from Brinkal, make sure to listen to the full episode. And don’t forget to subscribe to the Salesforce Admins Podcast. Podcast swag Salesforce Admins on the Trailhead Store Learn more Generative Canvas demo video Help Article: Visualize Your Data with Generative Canvas (Preview) Release Notes: Generative Canvas (Preview) Admin Trailblazers Group Admin Trailblazers Community Group Social Brinkal on LinkedIn Salesforce Admins on LinkedIn Salesforce Admins on X Mike on Bluesky social Mike on Threads Mike on X Full show transcript   Mike Gerholdt: This week on the Salesforce Admins Podcast, we're diving into the future of AI-powered customization with Brinkal Janani. Now Brinkal is a product manager here at Salesforce and he's leading the charge on Lightning App Builder and AI-generated apps. Today, specifically we're talking about Generative Canvas, forward-looking statement. I bet it's going to get renamed. So we're going to call it Generative Canvas for now, but literally watch the video that's in the show notes. This thing is so cool because it's going to reimagine how admins, how our users can interact with data and build dynamic experiences. And my two most favorite words, without code. Now, before we jump in, I want to make sure you're subscribed to the Salesforce Admins Podcast. That way when a new episode like this drops, boom, you can listen to it. I don't want you to miss out, so be sure to pay attention in whatever app you are using to either press that follow or subscribe button. So with that, let's get Brinkal on the podcast. So Brinkal, welcome to the podcast.   Brinkal Janani: Thanks, Mike, for having me.   Mike Gerholdt: Well, we're fresh off the heels of TDX and with AI and everything going on, I feel like the metaphor I've used of how fast technology changing is jumping out of a plane, it's moving very fast. I feel like it's jumping out of a plane and skydiving superfast towards the earth because with AI, everything's changing. And we're going to talk about some of the really cool stuff that you're working on on the platform, but let's get started with just learning a little bit more about Brinkal and what you do at Salesforce. So why don't you tell us what you do and some of the stuff that you work on?   Brinkal Janani: Sure. Mike, as you guys know, I'm Brinkal Janani and I've been at Salesforce for a little over than nine years now, and throughout my career at Salesforce. I've played various roles. I started my career as a software engineer in test, eventually transitioned to full stack software engineer, and now I'm a product manager overseeing a couple of product portfolios, namely Lighting App Builder and generating apps using AI. And that's where my focus is right now.   Mike Gerholdt: Yeah, I feel like everybody at Salesforce working on AI stuff, right?   Brinkal Janani: Yeah. And I'm glad to see whatever we're building at Salesforce is for the better. So I'm glad that to give the analogy, but that also means we are doing something that's fast-paced and would provide incremental value to our customers.   Mike Gerholdt: Absolutely. So you want to talk about, and this is where I will insert forward-looking statement because I feel like this is probably going to change names. So as of this recording, it's currently known as what, Generative Canvas? Is that right?   Brinkal Janani: That is correct, yes.   Mike Gerholdt: Okay. Let's talk about as known as Generative Canvas right now.   Brinkal Janani: Yeah, so I'm glad to talk about this project in particular because there's a history to it. I don't know if most of you know this or not, but I'm familiar that admins love concepts such as Dynamic Forms, Lightning Pages and Lightning App Builder. I mean, those are some of the popular features that admins love to play and use. And one of the fun fact that I wanted to also share was I was one of the original engineers in the team who built Dynamic Forms when it went tiered. So I have quite a bit of understanding on the expectations that admins have from the no-code tools like Lightning App Builder and the experiences they ship to our end users. And for years I've seen Salesforce admin utilize technologies such as Lightning App Builder, Dynamic Forms, even Page Layouts to control how the UI appears, period. And these no-code tools are the best and have been the best in what they do, but it's also nearly impossible to anticipate the full spectrum that the end users would need in terms of data and workflows, which is where Generative Canvas as a technology becomes really important. It unlocks a whole new way of interacting with data and workflow. So the general concept of Generative Canvas is, it goes beyond static user interfaces to more interactive and dynamic ones where you're talking to an agent and having those responses stored as Lightning components on the layout. That's the crux of it.   Mike Gerholdt: Yeah, this is one of those where the audio podcast has limitations, but I was watching the video and I'm thinking what you're working on is so far into the future because where we're at now with agents and admin, building agents is we add the agent, enable them, and they can be anywhere in Salesforce and we just ask them a text question. But this is actually really building a visual, well you call it a Canvas, a visual Canvas of chats plus also text and documents, right, meeting notes?   Brinkal Janani: Yeah, that is correct. I think what would really help understand what the concept is, it's a good example. A good example that we are starting off from is tackling meeting preparation use cases. We all have been in situations where we have back-to-back schedules where we don't even get enough time to see who we are meeting next. And I think that's true for most of us who will be listening in into this podcast, or have been in this situation. So imagine a sales executive who is handling multiple accounts and is literally in back-to-back situations with meetings when it comes to talking to these accounts and customers. And what it really is, and it's what you've seen with our customers, is like a very personalized approach, a personalized pitch. And executives not having enough time to prep that pitch, or even prepare for that meeting, has been the biggest disadvantage. And even if they do get time, they usually go to multiple UIs or even leverage multiple tools to be able to create that insight, which is why this was the first use case we felt is the best fit to tackle with Generative Canvas. Because Generative Canvas will make it super easy for any executive starting with sales executives to focus on the job to be done, and not a tool. And the reason I say this is because it's a single-page application where they are interacting with the agents on the backend. And as I said, agents are responding via text, via components that you put and organize on the UI. And not just that, you can personalize this layout or UI by moving these responses/components around on the Canvas and even resizing them so it exactly fits the way you would have imagined the UI could look like and carry this UI into the meeting so you have something to talk about, you have talking points to break the ice, or even you have insights if you want to cross-sell or upsell any of our products. So I think that example really nails the value of what we are trying to achieve with Generative Canvas. And obviously it's a start and we are going to grow from this use case to multiple other use cases, but hopefully this shows a value.   Mike Gerholdt: So I think the really cool part is you're thinking of quantitative and qualitative data because a lot of that, there's quantitative data, there's stuff that we can actually see in Salesforce. You brought up the sales example like number of opportunities or sum total of opportunities on this account. But the second part where your Canvas brings it together is all the qualitative data, which is all of that information, the chats, the extra documents, the insight that people need when they walk into a room to have that deeper level conversation as opposed to just the data that's in front of them.   Brinkal Janani: That is correct, and which is why I also feel like the future experiences is going to be both like a static experience and a dynamic one. I foresee living us in a very hybrid world where technology such as Generative Canvas will exist and coexist with technology such as Lighting Pages. To your point on Canvas, it's beyond static data. You get these insights and summaries that AI is able to generate and piece from the vast pool of data that we have in Data Cloud and at Salesforce. And that's the essence of it. Not just that, but once we start building in and start pulling data from public domain, you should be able to also get that data on the Canvas along with this data that's stored in CRM in your org.   Mike Gerholdt: Just help me elaborate on that. What would be data pulling from the public domain? What would be example of that?   Brinkal Janani: Example would be I'd like to learn more about the competitors of my current account. What is people, what is accounts, and what are they doing? So using that data, just having that competitive analysis and the most present one, which can only be learned by pulling the data from the public domain, the sales executive can use this information and potentially create a cross-sell pitch or upsell pitch for the existing customer. Just like having that lens, having that view inside can really help them create a personalized pitch for the customers.   Mike Gerholdt: Do you envision, and this is all me just thinking like, oh, this is kind of cool. So once somebody creates a Canvas, let's say for an account, would you envision that they would go back to that and then of course it would live update? Because obviously if you're pulling in news articles, it could do that. Like say six months comes down the line and the customer's up for a renewal, you'd want to go back to that same Canvas that you used to close the deal. Right?   Brinkal Janani: That's a great point, Mike. And I think one thing that I also want you to touch about Generative Canvas is that this Canvas is, as you said, are persistent, which means once you create this Canvas, they are there, they're part of your org and you can obviously revisit them. And obviously since Canvas is based on data, both static, and data coming from public domain, you would want to make sure they're still relevant, right? Because even the record details might have changed, a number of little records might have changed, a whole lot of data. The data that existed when the Canvas was created might be completely different from what exists right now. So having an ability to refresh the Canvas entirely would be essential to keeping Canvas not just persistent, but also relevant.   Mike Gerholdt: Yeah. Just on a non-Generative Canvas note, you mentioned in your intro that you'd also worked on Dynamic Forms, which I love to demo because I'm from that period of time when you couldn't have Dynamic Forms. Well, you could. My Dynamic Form was two different page layouts and a workflow rule to flip the page layout. That was my faking Dynamic Forms. What got you into the visual part of working in technology? I mean, it's obviously something you're really good at and something you're really passionate about.   Brinkal Janani: So I think I need to share this. When I was interviewing with Salesforce, I was actually interviewing the day right before my wedding.   Mike Gerholdt: Whoa.   Brinkal Janani: And the only reason I did this, and I'll never do this obviously again-   Mike Gerholdt: Well, I hope you don't have to get married again, in case your wife's listening.   Brinkal Janani: ... is because I knew what Salesforce was doing and what it continues doing, it's like creating these products which also create this community and just uplevels a whole lot of folks with the career insights and the career paths. And that is huge and that really resonated with me. And ever since my career at Salesforce, I've played multiple roles. My focus has always been delivering value for admins through no-code tools and specifically in my case, it's Lighting App Builder. So being an engineer or being a product manager, my focus has been how do I help admins unlock value for the end users through UI using no-code tools? And that has been my problem statement from day one at Salesforce, and which is why I'm super exciting to see how the feature's set, how the technology's evolving from beyond the static layouts in terms of Lighting Pages to a more dynamic world where everything is personalized, everything's AI driven. But as I said earlier, I also feel like the reality, the future is mostly hybrid with both technologies coexisting seamlessly.   Mike Gerholdt: When you say hybrid, do you mean humans and technology or what do you mean by hybrid?   Brinkal Janani: When I say hybrid, I think I see a world where it's not completely one-sided where you only have static experiences or you only have dynamic experiences. When I say hybrid, I mean I see a world where you have both kind of experiences, probably start from a static experience to a Lighting Page, and eventually transition to a more dynamic experience through technology such as a new Canvas where you're conversing with a tool, where you're conversing with an agent, and updating the layout on the fly.   Mike Gerholdt: Yeah, okay. And everything turned out okay at the wedding, I'm assuming?   Brinkal Janani: Yes.   Mike Gerholdt: Okay, good.   Brinkal Janani: Safe to say that.   Mike Gerholdt: Well, we know the job thing worked out because you're on the podcast. So I don't know I've ever had anybody on that interviewed before their wedding day. But good for you. When you sit down and think of all of the stuff that AI has to interact with now, and what you're working on for this Generative Canvas, I'm not an engineer, but it's very easy to get caught up in the here and now. How are you trying to plan for two to three year out technology that you might not even know exists and pull that in for the next generation of Generative Canvas? I mean, you have to know you're building something out of Minority Report, right?   Brinkal Janani: And I think that's a great question again, Mike. To be honest with you, in this period of time, it's extremely difficult to even look out two years out in advance just because the pace of technology is changing so drastically. But one thing that remains constant, regardless of what period we are in, are the problems that you want to solve for your customer base. Those are not going away. Eventually technology needs to be able to solve customer problems. So my focus always has been less on the technology itself, but figuring out the right problem set to solve in the right period of time using the right technology, and that's what I want to achieve. And that's what I've always been trying to achieve. Two different sets of technologies. So eventually in time technologies might change, but you still have the same problem set to deal with. And solving some problem sets might become just easier as in when technologies evolve. And I think we just need to keep on revisiting that list of problems that you have that you want to solve for your customers.   Mike Gerholdt: Yeah. Last question, because I threw some hard ones at you, but you also came with the wedding thing. That's pretty unexpected. When your... I mean, man, this is so, "In the future we're going to have flying cars, but we don't know." When you're planning through and looking for inspiration, this is the one thing that I think I learned it from a long time ago, I had a guest on the podcast that talked about board games and how they found inspiration in board games when they were creating a product in Salesforce. Because you work in very visual products, where do you go to find inspiration for ideas like for Generative Canvas that you're working on?   Brinkal Janani: That's a tough question again, Mike,   Mike Gerholdt: I don't ask easy ones. Nobody comes on for easy questions.   Brinkal Janani: All right. Point taken. I think the answer is two-folded, one is very personal and one is more cooperative. I'll start with the personal one. I'm also father to a three-year-old daughter.   Mike Gerholdt: Oh, boy.   Brinkal Janani: And it's been quite a journey, and mostly good. The reason she is an inspiration is because it's so amazing to see how a creature of such small size can learn and absorb from visual experiences, from sensory experiences, so quick. And depending on what the experience is, the message that they learn is drastically different. And I know it seems like a very far-fetched connection, but if you tie the dots, it's actually not that far. That is my daughter has been an inspiration just to keep my mindset more agile and rapidly adopt with changing environment and learn from it and keep on delivering value. So that has been something that I'm really grateful for. The other thing is a lot of people, we have an amazing set of people in the world right now and everybody's doing something very amazing. So this thing connected, and in the know on what's happening around you, also serves a good idea for inspiration.   Mike Gerholdt: Yeah, no, I mean think a lot of us are still wrapping our heads around... I have a good friend of mine has a young child too, and he told me the other day, "He just ran up to the TV and started touching it, expecting it to do things," much like his phone. And that to me is like, whoa, because that's the world they live in. As a parting gift, if admins are definitely thinking about Agentforce, and AI, and everything that they can do in Salesforce, what would your best advice be for admins around getting ready for just AI and Agentforce and learning this new world that we live in?   Brinkal Janani: I've always felt for Salesforce, the local community groups that we have, the Trailblazer communities that we have, Trainforce, TDX, and just like one-on-one engagement with product managers and MVPs have been a very solid ecosystem that has helped spread the knowledge and just up-level everyone by sharing knowledge, talking about the problems, and talking about how do we unlock, or solve for these problems using the technology that the Salesforce has right now. And I think it will be vital for us, especially product managers at Salesforce, being part of those communities and local groups and talking about the technology, the product that they're working on, and helping customers connect with it. I think that would be a key, not just for success of products, but also for success of our customers.   Mike Gerholdt: Yeah, I couldn't agree more. I mean, it can be oftentimes very isolating to sit in your office and feel like you must be the only person working on this right now, and you're not. And getting out to a user group and hearing stories and thinking, "Oh, you just need to change a couple terms, but we're working on the same problem," and seeing different approaches to things is always very helpful. Brinkal, thanks so much for being on the podcast. I can't wait to have to go back and edit this once we rename Generative Canvas three or four more times.   Brinkal Janani: And yeah, thanks Mike for having me. This has been special.   Mike Gerholdt: No, it'll be great. So that was a fun conversation with Brinkal. I'll be honest, I don't know of any podcast guests that may have interviewed the day before their wedding. Usually there's things to do. You know what, Brinkal is like, "I'm going to work at Salesforce and build the future of applications." So I'm so glad everything's working out for him. But man, I can't wait to see your reaction to some Generative Canvas stuff. Again, I'll include all the links in the help and the video to watch it. If it works for you, go for it. I think they're going to roll it out even farther, forward-looking statement. But this is going to be cool. If you enjoyed this episode, hey, do me a favor, share it with a fellow admin. You can tap the dots in Apple Podcasts and click share episode and then that way you can post it on whatever social platform you are on or, text it to a friend. Or share it in your community user group. I'd be a fan of that. Of course, all the resources, including the transcript, are in the show notes for this episode. And where are those? Well, those are your one-stop shop for everything admin, which is admin.salesforce.com. Now don't forget to join the conversation over in the Admin Trailblazer group. Don't worry, the link is in the show notes. Where's the show notes? Admin.salesforce.com. Look at that. It's like we thought about this ahead of time. I promise we did. All right. Until next time, we'll see you in the cloud.  

Today on the Salesforce Admins Podcast, we talk to Kurtis Kemple, Senior Director of Developer Relations at Slack. Join us as we chat about what’s possible when you combine Slack, Salesforce, and AI agents. You should subscribe for the full episode, but here are a few takeaways from our conversation with Kurtis Kemple. Starting at square one in tech Kurtis’s path to his career in tech is truly inspiring. Not only is he a completely self-taught programmer, but he learned those skills while incarcerated. It was hard to get any sort of job when he got out, let alone convince someone to take a chance on him as a software engineer. Today, Kurtis is the Senior Director of Developer Relations for Slack. His role is primarily focused on advocacy, with a focus on improving the developer experience through thoughtful product design and community input. So he’s the perfect person to talk to about what’s possible with Slack, Salesforce, and AI agents. Slack is the OS for work When it comes to collaborating with your team, Kurtis sees Slack as the OS for work. It’s a space to bring together everything you need—your communications, your documents, your data—all in one place so you can start getting things done. Switching contexts can be a productivity killer. That’s why Slack’s integration with Salesforce is so powerful, because it allows you to have everything right at your fingertips without needing to go back and forth between windows. Whether you’re looking at Salesforce data in Slack to have a conversation with a co-worker about an opportunity, or updating your team on what you’re building in Salesforce, seamless authentication means you can do everything from wherever you happen to be working without having to switch back and forth. Agents and automation inside Slack The possibilities are even more exciting when you throw Agentforce into the mix. As Kurtis points out, Slack actions are part of the list of standard actions. That means you can build custom agents that use data from either platform to launch workflows, run a quick analysis, and much more. Kurtis also gets into how you can customize Agentforce by plugging in various LLM libraries, or connecting it with external services or authentication providers. As he explains, prompt templates are powerful tools for controlling your agents’ responses so that they fit into your business processes. This episode is a deep dive into everything you can do with Agentforce and Slack, so be sure to take a listen. And don’t forget to subscribe to the Salesforce Admins Podcast to catch us every Thursday.   Podcast swag Salesforce Admins on the Trailhead Store Learn more Blog: Getting Started With Slack and Agentforce Integration Trailhead: Connect Your Agentforce Org with Slack Trailhead: Learn How Agentforce and Slack Work Together Salesforce Admins Podcast Episode: What Can Salesforce Admins Do with Slack Integrations?   Admin Trailblazers Group Admin Trailblazers Community Group Social Kurtis on LinkedIn Salesforce Admins on LinkedIn Salesforce Admins on X Mike on Bluesky social Mike on Threads Mike on Tiktok Mike on X Full Transcript Josh Birk: Hello, Salesforce admins, your guest host Josh Birk here. Today on the show, we're going to welcome Kurtis Kemple, who's overrunning developer advocacy at Slack. We're going to talk about Slack, we're going to talk about AI, we're going to talk about Kurt's beginnings, and this was a great interview. Very happy we got it on tape. And so, let's go right over to Kurt. All right. Today on the show we welcome Kurtis Kemple of Salesforce Developer Advocacy. Kurt, welcome to the show. Kurtis Kemple: It's a pleasure to be here. Thank you so much for having me. Josh Birk: So let's start, I usually start talking about people's early years. And you have a very, let's not say unique, but a very interesting early days of getting into computing. You're self-taught in web development and you learn that in prison. Correct? Kurtis Kemple: Yeah, that's correct. I prefer to think of it more as self-guided. I'd like to take all the credit, but I was still reading books, watching YouTube videos, paying for courses. So you're still creating a learning environment more so than just I had... Well, in prison I did have nothing, but then I just opened a laptop and just essentially went at code with nothing. I don't want to essentially leave out all the people who I wouldn't be here without them. Josh Birk: Did you have an interest in computers beforehand or was it just sort of a, "Here's this moment in my life and I want to make changes, and I want to learn something new"? Kurtis Kemple: Well, it was definitely that second one, but not related to code. So it was pure happenstance. Prior to incarceration, I had very limited experience in technology. I had a pager once when I was 18, cell phones were coming out, and I would download music off LimeWire. That was it. I had to type with two fingers, like pecking at the keyboard. Josh Birk: Did the looking over the keys with the finger on either side of it? Yeah. Kurtis Kemple: Oh my gosh, absolutely. Absolutely. Josh Birk: I also love doing interviews like this because we always accidentally date ourselves. Kurtis Kemple: Oh, my lord. Yeah. Josh Birk: In some wonderful way. Kurtis Kemple: I know people are going like, "What's a pager?" They're like, "Oh, I've seen those on TV." Josh Birk: I know, right? "Did they beep or something like that? They had numbers on them?" Yeah. Actually, I didn't get a cell phone for a long time because my dad's a doctor and he had a pager, and I remember hating the fact that he would be taken away from the dinner table because his pager is on his belt. I'm like, "Oh, I don't want to have constant access to people always being able to call me." And now, of course, I'm the dude that my wife slaps my wrist because I'm picking up the cell phone during dinner with friends. Kurtis Kemple: Absolutely. Josh Birk: Just as addicted as everybody else. So tell me a little bit, 'cause tech is not the easiest to break into, and there's ageism, there's sexism, there's racism, there's the old white guy syndrome, all this kind of stuff. With a record of a conviction, what challenges were you facing there? Kurtis Kemple: Sheesh know, I'd say that- Josh Birk: How many hours do we have? Kurtis Kemple: Honestly, honestly. Look, this is for all you PMs out there, I'm going to throw in a little product analogy, but there's a ton of upfront friction. It's like they would never let me get through to the other side to realize the value. So it's like having that record, it's immediately, oh, that's like, "I don't want to mess with that." They had no idea that on the flip side of that, I was willing to work so hard to prove myself. So hard. And so, I think there's two sides to every coin, but it was a big blocker. It's just immediate nos. You have conversations with people and they're like, "Oh, that's interesting." And then they find out about the record, or back then you had to check a box, "Have you been convicted of a felony in the last seven years or since you've got out?" And it's like, "Yeah. Yeah, I have. But I thought the whole point of going through half a decade of incarceration was that I did that. I paid my dues." But no, you're doing time for forever after you get out in the US, honestly. Josh Birk: Yeah. Well, I am very glad that you have managed to break your way into tech and join us. Tell me a little bit about your current job. Kurtis Kemple: Yeah, so from there all the way learning and coding, I went through an entire career in software engineering, and then I ended up making a switch to DevRel about eight years ago. And then I went from IC up in DevRel, really, up to leadership. And so now, I'm senior director of developer relations here at Salesforce, and I focus on Slack, Slack platform. Josh Birk: And tell me a few of those highlights, because you've worked for companies like, if I'm remembering your CV correctly, like Amazon. Correct? Kurtis Kemple: Yeah, I've worked for a ton of pretty big companies, some that might not be as well known, but you'd know them in certain verticals. So I worked at Integral Ad Science, which is actually, if you've ever seen an ad, they're the analytics platform behind the analytics platforms. So it was a really big one. And then from there, a bunch of places that culminated with my last job as a full-time engineer. I was the tech lead of the UI team at Major League Soccer. And so we built- Josh Birk: Cool. Kurtis Kemple: Yeah, it was a really fun one. So we built all the interfaces that fans interface with, like user interface, literally, chatbots, Roku, Fire TV, websites, mobile apps. You name it, we built it. Josh Birk: Nice. Very, very nice. Now, this is kind of a personal question for myself because it's one I get frequently, and I apparently struggle to answer it correctly because I still have people in my life who have no idea what I do. What is developer evangelism or advocacy? Which of the labels or do you land ,on evangelist or advocate? Kurtis Kemple: So, I think at any point, you might be doing a bit of both in reality in these days, but if we wanted to put a label on it, you could say evangelism is more outbound-driven. Advocacy, ideally you're doing a lot of inbound. We're advocating, both internally and externally, we're providing feedback. I would say I lean heavily, I'm a very product-driven person, so lean on the advocacy. I want to unlock better developer experience through improvements to the product. Josh Birk: Got it. Kurtis Kemple: That's my main thing. But also, I love getting out there and just talking [inaudible 00:06:54] and engaging with people, interestingly enough, because I am introverted, but I love it. I absolutely love it. Josh Birk: It's such an interesting trend in advocacy and evangelism. I remember having a beer with a former evangelist actually, and I'm like, "Well, you know I'm a closet introvert. Right?" He's like, "No, you're not." I'm like, "No, dude." Kurtis Kemple: Yeah, they have no idea. Josh Birk: "You have no idea." One of the first things I told my therapist is, "I have raging social anxiety, and so I decided to go into public speaking." Don't ask me why. Don't ask me why. Kurtis Kemple: "I wanted to live my worst fear daily." Josh Birk: Right, exactly. Exactly. Oh, yeah. Okay. So our focus is admins, and what I'm kind of curious is, because first of all, I'm a huge Slack fan. And it's one of the purchases that I'm like, "Oh, I get this one." Sometimes Mark buys a company, I'm like, "I don't know who they are. I don't know why we would use them." First of all, I'm very glad we purchased Slack because it finally answered an age-old question of which IM service should we use to talk to our teammates? And yeah, finally have a- Kurtis Kemple: Really? Josh Birk: Oh, it was horrible. It was G-Chat and it was Top This, and then there's Chatter, and to have everybody agree was great. But how do you see, in the Salesforce ecosystem, what is Slack's role as a product? Kurtis Kemple: Yeah. And this is so great 'cause this ties right into administration as well. It is the operating system, without a doubt, that is, I don't know, I think of a computer, extrapolating out a bit higher level. It is aggregating all of the useful tools, access to information, communication channels, but that is very broad, very general. It's covering it for everything. Slack narrows in onto one particular aspect of your life, work. Or actually, I won't even say work, but I will say collaborative organization and communication, 'cause we have millions of communities on Slack as well. And yeah, I think that it's about that. Honestly, it's that. It's an operating system to help you organize and be productive and get things done. And especially now when you put the lens of work on it, it becomes even more valuable. That starts turning into dollars. And it's tangible, tracks billions of dollars. Josh Birk: Yeah. Yeah. And I really love positioning it as it's kernel, is kind of the... Oh gosh, I just made an OS pun when I didn't mean to. Kernel is sort of that collaborative tool, but then there's all these other layers to it to help people be more productive and be more collaborative. And over the years, we've had tighter integrations with Slack and Salesforce. What are some of the key integrations that come to mind when I'm using Slack, but I'm also trying to interface with Salesforce? Kurtis Kemple: Salesforce channels. Without a doubt, if those are not set up, oh my Lord, it is a game changer. So quick links at the top, taking action within Salesforce. Again, for me, it's like the work operating system, I don't want to switch context. I don't care if it's to switch a window for three seconds to click a button, or I have to go out there and do a bunch of other things, it breaks the flow of work. And to me, what I'm actually most excited about is when you look at a Salesforce channel, you've got your conversation. Let's say it's related to whatever, like an account, there might be your opportunities, or the health score or recent notes, and all this stuff is right there in the team channel as soon as you drop in. And I'm sure that we use Slack more than most people use Slack 'cause we are Slack and Salesforce. But if you are starting to take advantage of those tools, I've got all my documents related to the DevRel team, in our Dev Rel channel. It's just so convenient. But when you integrate other systems in that, like Salesforce, and now I just have a canvas, but it's real time updates and back and forth collaboration. Okay, my job keeps me mostly in Salesforce platform. A, it's day to day business as usual, but if I spend most of my time in Slack, I'm no longer breaking the flow of work, even if it is just a few times, that can really have an effect, compounded. So I would say definitely that. And now, the counterpart to that is we've got the Slack channels in Salesforce. Josh Birk: That's so cool. That's so cool. Is that GA? How do people turn that on? Or is it just there? Kurtis Kemple: It is GA. How do the people turn it on? I think it's just there if you already have things connected. So if you set up a Salesforce channel connection, it will show up for you in the Slack, in the Slack UI with the Slack way of doing things. And if you're in Salesforce, then you get the Slack channel, messaging still available, everyone conversation one place, but you have your Salesforce UI that you've custom designed and built around you so that you can do your work more efficiently. I'm sorry, I swear I'll let you talk right after this, but you hit on something so key, I swear to God. Josh Birk: No, go, go, go. Kurtis Kemple: It took us a while to get here. Look at this. We're like three, four years into this thing. More, more. But when you look at how to thoughtfully integrate something like Slack and Salesforce, there's going to be some trial and error. And I know people wanted it faster and I wish we gave it to them, but I'm actually just happy that the things we're building now are so much in the flow of work. There's some things coming up that I don't know if I can talk about. I should have ran through the list, but I cannot wait. Cannot wait for folks to see 'cause bridging Salesforce and Slack in ways that are just amazing, and are really going to close the gap for ease of use and just being able to stay where you're meant to be and do your job. Josh Birk: Nice, nice. I want to touch on that a little bit, but I also want to kind of applaud the engineering team over there, because the demos I've done, and I'm old, I've done integrations. My first application was a CGI, getting things to work, but integrating systems sometimes isn't easy. The seamless OAuth between Slack and Salesforce is pretty cool. The idea that you have this mirror user and then you just tell Salesforce, "This is my user over here, go be Slack." And then I'm like, "What do you mean there's no OAuth window? What do you mean I don't have to worry on tokens?" Kurtis Kemple: You're already you. You're already you. Josh Birk: Yeah. Kurtis Kemple: Yeah. Josh Birk: Yeah. Did you have a follow up there, or? Kurtis Kemple: I do. I do, I do, I do, I do. Yeah. So Seamless OAuth I think is foundational. As a matter of fact, I see this as foundational to the entire Salesforce ecosystem. And so now that's seamless, all right, so here's the deal. Imagine now it's like, oh, click of a button, I deploy a Slack workspace for my Salesforce org. That's awesome. But now, it's like I want to also extend my operating system, my Slack org, so I want to start building and deploying apps. Well, what if that user was also seamless off to Heroku and you could just deploy your apps immediately to Heroku? And what if you wanted to use MuleSoft APIs to further extend it? But guess what? You're the same user there, too. So now, it's like you can immediately spin up these really ridonkulous, enterprise-grade-ready, all seamlessly authenticated systems and just go to town. Josh Birk: Right. Well, and that really starts to put the cap on the concept of it being an operating system. Kurtis Kemple: Bingo. Josh Birk: It's Slack, it's Salesforce, it's MuleSoft, it's Heroku, it's whatever in our ecosystem that you have to touch. And I'm going to be very honest, when canvases really became a thing, I was kind of like, "Okay, I've got docs, I've got this kind of stuff." And then our team recently tasked me with trying to track some pilot features, and I was struggling with, "Well, where should this live?" Because of course, this is not going to surprise you, my first instinct was, "Custom object, throw it in the enterprise org that we all collaborate on, keep them as records, blah, blah, blah, blah." Kurtis Kemple: Yeah. Yeah. Josh Birk: And then- Kurtis Kemple: Permissions, groups, we're all good to go. Right? Josh Birk: Right. And then I thought about exactly what you were talking about before, how many touch points is that for somebody to go and read something that they're not... This is something people are intended to scan. Kurtis Kemple: That's it. Josh Birk: This isn't a research doc. It's a "Here are some things to think about in front of Dreamforce." And I'm like, "That's why I put it into a canvas." Kurtis Kemple: That's it. Josh Birk: It's right there and you can just click that button and go. Now, speaking of OSs and conversational UIs, because I think this touches a little bit on what you just said about deploying workspaces and things like that, because we now have Slack actions in Agentforce as standard actions. Can you tell me a little bit about those? Kurtis Kemple: Yes, I absolutely can. So, you want to be able to take action in Slack from Agentforce. It's like if that is the work operating system and agents are truly supposed to be doing work on my behalf, I want those reports summarized and put into a canvas and added to my channel. I want to set up DMs maybe with somebody from customer success and sales, or IT and wherever. I think as we see Agentforce start to grow more in maturity, we're going to need to see a lot more Slack actions. And fun fact, like little shout out to the DevRel team over here. If you want to extend beyond the Slack actions that exist, we've already got the docs all set up for you for building custom Slack actions for Agentforce. So you can literally just, world is your oyster with Slack. And technically, they work for any service. We're not doing anything particularly bespoke. But it's nice, because you can set up a Slack app which gives you all the proper authentication to either be you or a bot, depending on how you want to set all that up. And then you can now start taking that even further. You can have agents doing things on your behalf, like responding to messages, or summarizing things and aggregating stuff, updating records in Salesforce, or updating channels and canvases and lists within Slack, kicking off Slack work flows, all kinds of stuff. But to be honest, that's all very well and good, except I think until we get agents more deeply integrated into the UI of Slack, into your flow of work, we're just scratching the surface. And we're working on some cool stuff now that you'll see, but right now it's really just the agents are there, you know what I mean? We brought them into Slack, but they're kind of side card, they still have access to all your info, have access to the relevant channel and stuff like that, or channels. And now, you can actually message them in channel, which is a big step forward, but that's still very much the initial phases. I can't wait for all this stuff we'll be able to do. Yeah. Josh Birk: Okay. I'm going to free you from the restraints of a forward-looking statement. Kurtis Kemple: Let's do it. Josh Birk: And we're going to walk right into the land of theory. So everybody listening to me right now, this is not a product roadmap, Kurtis is not promising you anything. I am just going to ask what's your dream version of an agent residing in Slack? Kurtis Kemple: All right, so I think it's twofold. Oh my gosh. Oh man, this is going to get so deep. Here we go. Josh Birk: I love it. Kurtis Kemple: Number one, I think agents should be able to be preferential. They should learn about you over time, specifically. Not general knowledge, not learning about all of the people in Slack, you specifically. So, I would like to see more of that happen because then you get into generative use cases. It would better understand what actions to take, better understand what things to do because of you and how you specifically have interacted with it. So I think that's one. And then two is like, this is going to sound so funny 'cause we're the conversational interface, but it's like we've already cracked that nut. We're good there. What if agents were more in flow into literally to the micro flow of work? When I highlight texts, I should be able to kick off agent actions and send that to agents. And then I would love to see more eventually agent orchestration. So right now, we've got a few agents, but let's assume you're like X-Corp, 5,000 to 10,000 employees. You open up Agentforce for them, they build one agent per, not everyone would, but realistically, now you're talking about thousands of agents within your environment. That could be a lot to handle. So I think winning the game there is about making it a lot easier to synthesize and have AI help you manage that. Josh Birk: A couple of meetings ago we were joking about whether or not we should consider adding a mental health topic to our Agentforce. And as you're saying that, I'm like, okay, so one of these days my Slack channel is going to have an agent in it and that agent's going to be like, "Dude, you've got too many meetings this week. Go take a walk." Kurtis Kemple: Exactly. Exactly. Josh Birk: I love it. I love it. Kurtis Kemple: Or there's like, Hey, there's 13 agents in here and going through their topics and descriptions, their classification, these are all doing roughly the same thing or something like that. Or, "Here's all these agents around your org that are all doing the same thing." I think a lot of stuff, I don't know, I've just been digging AI lately in, one, non-conversational use cases, so kicking off from buttons and stuff like that. I think we should get a little bit back to that. And then two, I would like for agents to not just be specific to a topic or an industry, but specific to me. Josh Birk: Got it. I love it. I love it. One of the things I preach often is if you are a Salesforce admin, you're already an AI builder. If you know Flow, you already know how to extend an AI. Kurtis Kemple: Absolute. Josh Birk: [Inaudible 00:21:56]. And one of the things that I've seen some of your use cases, and that's what I want to dig into next, but it's interesting how simple Flow can be, but also how much more power it can add to an AI, into an agent, simply because of its ability to integrate with external systems and other things. What are some uncommon use cases that you've tinkered with when it comes to... And we'll do Flow/Apex because they're pretty parity when it comes to features these days, so don't limit yourself to just Flow. But what are some uncommon use cases where maybe somebody's like, "Oh, I didn't think an agent would do that"? Kurtis Kemple: When we think of Agentforce, we immediately just zoom in on that is our agent, we have the reasoning engine, and then we have the actual agent itself, and that's the only access to LLMs within Agentforce. And that's not true. So, Salesforce has the LLM Open Connect standard. You can attach any LLM you want to as a foundational model if you have a data cloud. And you can start doing models for all kinds of things like specific actions, like a model per action. This was kind of blowing people's minds when I started showing them this. I have actions that can help me summarize text, actions that actually help me review and rewrite text, actions that can help me with code instruction, actions that help me analyze data. I have my Salesforce developer advocate model that helps me learn Apex Lightning web components, policies and setup. I'm becoming an admin myself. I'm starting to get pretty dangerous in there. Josh Birk: I love it. Nothing wrong with being a double or a triple threat out there, that's for sure. Kurtis Kemple: Absolutely. Josh Birk: Yeah. I love it. I love it. Kurtis Kemple: I could dive more into third-party services too though, 'cause I think that's a big one that we've not tapped on. Josh Birk: Go for it. Kurtis Kemple: Yeah, yeah. So aside from custom LLMs, which again, now you can handle any use case you want, but Salesforce is the most integrated platform probably in a history of platforms. And I think people try to leave a lot of that at the door. And so we'll get into Flow and Apex next, but those are the gateway for this. Because I can set up external services or auth providers, or like I said, custom models, and then all of those things, prompt templates are great for really controlling the responses in Agentforce. All these tools become available to you. And so I have an agent, engineering agent that... Actually, this is a cool one. It pulls down my PRs for me that I have, and it will spin up GitHub Codespaces. And I can run the code, review it, and then be like, "This looks good. Please break that down, merge this for me." And it goes off and does it. Josh Birk: I love it. I love it. And for the admin, non-technical group PR pull request, so basically we keep coming back to this theme. We are your engineers working? Well, they're probably talking to each other in Slack. They're probably talking about that pull request in Slack. So why get out of Slack once they've already approved it in Slack? So why get out of Slack in order to approve it and then add that to the code base? Kurtis Kemple: That's it. I'm in the flow of work. I never left Slack. I sat down and I said, "Oh, you know what? I have a few minutes. Let me"... Okay. Now, here's the beauty. I said, "Oh, I have a few minutes. Let me see what I can tackle." If I had had to leave to go to GitHub to do that, I would've never happened. I would've looked for things I could do a few minutes within Slack. Josh Birk: Yeah. Yeah, you know how some people have a swear jar. Right? Kurtis Kemple: Mm-hmm. Josh Birk: I thought about just putting a jar in the kitchen to put a penny in every time I walk into the kitchen and forgot what I was going to do. Kurtis Kemple: Oh my gosh. Josh Birk: So yeah, if I have to leave Slack, I'm probably getting a coffee or figuring out why the cats been away or something. Yeah, totally. Kurtis Kemple: Literally. Josh Birk: Totally. I want to dig a little bit more in the custom models, because that's actually one thing I have been curious about. I have not played with it much myself. And one of the things that keep kind of, when I'm talking about AI, is kind of breaking down this concept that it's a rocket science sort of thing. Yes, I've seen the math and it made my eyes bleed, basically. If you want to understand the formulas, you're probably going to be a Stanford research scientist. If you want to use the technology though, you already have the skills. And one of the things it's interesting about the concept of a custom model is when we do an API call to OpenAI, OpenAI is intended to know how to cook a rotisserie chicken. Admins don't need to know, our users don't need to know how to use a rotisserie chicken when they're in Slack. Kurtis Kemple: Yeah. I don't need a model that can talk like a pirate. You know what I mean? Josh Birk: Exactly. Although, prompts that talk like a pirate are a fun demo to do. Kurtis Kemple: They are great. Josh Birk: But what is involved in creating a custom agent, like when you say, "I created an agent that understands Apex and LWC"? Kurtis Kemple: Yeah. Yeah, so I actually created an action and I just attach that action to the agent. So here's a cool thing, I'm not training anything. AI has gone from frontier to common product so fast. If we look at the Wardley scale, probably faster than anything else in the history of humankind. And we are now at the stage where AI, you can have access to very specifically trained models to do something very well through a simple API call. I'm sorry, I shouldn't say simple, because even API calls can be complex for those unfamiliar. My point being, is that you don't have to train your own models. You don't have to be an AI or ML engineer. You don't even have to be a developer. If you can open up Postman and figure out how to use a couple API endpoints, that's all you need to become very active with AI. And then it's just a matter of identifying which models are best for which job. And trust me, there is no shortage of that information online. Josh Birk: Where are some of your favorite places to find custom models? Kurtis Kemple: I definitely lean heavily into Hugging Face. So it's basically like the GitHub for models and data sets and stuff like that. I like it because it's the platform I found that opens the most transparency. So a lot of them can even record the carbon footprint of what it costs to train their models. And they have little cards right there on Hugging Face. Yeah. Josh Birk: Oh, very cool. Kurtis Kemple: Yeah, it is. It's cool. You can also see the data sets for a lot of them, what they're trained on, but they also have access to all your big, what I call the 100 million club models. You need $100 million or more to train. They've got access to all those as well. So, I like it because then I use one API, the Hugging Face API, and then I just switch the model as one property and I can completely change the response. Absolutely. Josh Birk: Oh, that's very cool. That's very, very cool. So let's broaden a little bit outside of the world of Slack and Salesforce, because I know you've also talked in general about the importance of democratizing artificial intelligence. Give me a little bit of a pitch on that. Kurtis Kemple: All right. So, it goes back to incarceration. And this is also why I said self-guided, not self-taught. The only reason I was able to succeed is because enough people had come before me in tech and been willing to share what they had learned. That is not common. Go Google how to become a doctor. There's not doctors out there writing blog posts on how to become a doctor. There's not electricians doing it, accountants, bankers, I can go down the list. Now, there are some, don't get me wrong, I don't want to talk negatively about any industry. Trust me, there are people out there, but there's not the wealth of information that there is for learning tech, for especially coding. Look, even in a Salesforce ecosystem. So I look at Trailhead, I am so blown away by what I call the opportunity funnel of something like Trailhead in the Salesforce community. Say what you want, but nothing is a career creator vacuum for people who have non-traditional or resource-deficient backgrounds and have a viable career. So sorry, circling back. Imagine if I had had a small un-internet connected device with a model, trained specifically on code instruction to help me along in there, instead of one textbook and my teacher being willing to print out some pages on JavaScript MouseEvents. Josh Birk: Yeah. And I think a lot of that, I think there's a harmonious cycle there, because I think a lot of it is also how, especially techie people, whether they're an admin or a developer or an architect or whatever, I feel like we learn once we do it sort of thing. We can read it, we can see it, but the best... This statement alone is probably why I'm in computers. Back when I was in school and we got to play around with Apple IIs, I think they were Apple IIes, and our teacher said, "Do whatever you want in this class. You are not going to break the computer by typing code into it." And it was all of a sudden just like, it's like, "Okay, I'm going to go do go-to lines for 1,000. Let's just to see what happens." And it was so freeing, so freeing just to be able to be like... And JavaScript, and anytime somebody asks me, "How should I start?" I'm like, "Start with JavaScript." It's free. You can do it in a browser. You can instantly see the results when typing into it. It's great. And back to your point, my father was a doctor and I asked my dad how to become a doctor, and his advice was, "Don't." Kurtis Kemple: There you go. Right? Josh Birk: Yeah. Kurtis Kemple: Yeah. And I just think even for becoming a doctor, I think AI changes the technology landscape, the learning curve, the accessibility. The accessibility. So I think another piece of this too, and about democratizing AI, again circling back to the $100 million club, is that a lot of information is available through AI, but you know what's not available through AI right now, honestly? Expertise. Josh Birk: Oh yeah. Kurtis Kemple: Right? Josh Birk: Yeah. Yeah. I feel like we're still in that phase where we're the experts and AI is trying to catch up to us in a lot of ways. I've used ChatGPT just to write a simple Apex class, and sometimes it not only nails it, I'm like, "Okay, that's prettier code than I would've written." And then sometimes it's like, "You dumb bot." Kurtis Kemple: That's it. That's it. And it's literal, it's statistics, it's probability. And the reason why, is 'cause if it's trained on the whole internet, chances are there's as much bad info about it as there is good. Josh Birk: Yeah. Kurtis Kemple: But also, I think about this, even getting deeper than that, 'cause that's still recognition. Right? Josh Birk: Right. Kurtis Kemple: But you could tell me every song on a Taylor Swift album as AI, but you can't tell me why Taylor Swift arranged to the tracks the way she did, what she is listening for, when she knows it's the right time to make an album, if she even cares at all. And it's just like all is actually just like, "Woo hoo." Who knows? Josh Birk: Yeah. And I think based on how much Swifties themselves analyze her songs, that by the time AI can tell you what Taylor was thinking when she wrote a song- Kurtis Kemple: Too Late Josh Birk: ... we're at the singularity. Kurtis Kemple: That's it. That's it. It's too late. Josh Birk: It's over guys. Kurtis Kemple: Yeah, even borgs and everybody knows why. Josh Birk: Exactly. All right, so totally switching gears. I want to give you an opportunity to talk a little bit about Forthright. Kurtis Kemple: Oh, okay. So Forthright, actually, I do not run that anymore, the consultancy, but it was fun. It was amazing. So I was doing DevRel consultancy actually before I joined Salesforce and Slack. I had had a number of jobs in developer relations. This is still very awesome and relevant, but I just felt like I was... I don't like to do things kind of in isolation. I don't want to make assumptions or not have other information to lean on. It makes sense when we want to put care into something, we want to ensure we do it. It's actually how I got into DevRel. When I was at Major League Soccer, we wanted to adopt React Native and GraphQL. They were both being used in production by a lot of places, big companies and stuff, but really nobody was talking about it yet. Josh Birk: Interesting. Kurtis Kemple: So I selfishly, very selfishly started writing blog posts about them, talking to people from the companies. I started GraphQL NYC Meetup. I started doing talks at the React Native meetup, just to hear how other people were doing things so that I could do a better job at MLS. I was a tech lead. It's literally my job to understand the tech. That kind of just stuck with me as a really good and powerful way to learn. So, after having some time in DevRel, like Gatsby, AWS, like you said, Apollo GraphQL, I had my own thoughts about how it should be done, but what I realized is that I don't have enough information, so clearly I'm missing pictures. And secondly, I'm not going to be able to really provide that type of environment for myself in a full-time job, because it's one environment and you're in that ideally for years. Right? Josh Birk: Yeah. Kurtis Kemple: So I did consulting. I worked with startups, scale-ups, enterprise companies, and I just went deep into DevRel. Plus I'm, of course, interested in talking to everybody in the industry. And yeah, it really helped. I really feel like I had such a better understanding at the end of that year and a half. And so coincidentally enough, there was a bit of a downturn in the market, hot leads started turning into cold leads, and then right when I was like, "This is my least favorite part of the job, do I want to keep doing this?" I actually got the referral, like a reach out about an open position at Slack. Josh Birk: I love it. I love it. All right, Kurtis, one final question. What is your favorite non-technical hobby? Kurtis Kemple: Oh, I have to pick between two. I'm going to say photography and street photography. It's very close to that and riding my motorcycle, but- Josh Birk: I love it. Kurtis Kemple: Yeah, yeah. Both of them serve the same purpose though, which is what we talked about, anxiety and pressures and work and stuff. And with my history, it's very easy for me to live in the past or the future and not in the present. And both street photography and my motorcycle forced me to be extremely present. And so I really enjoy those two hobbies because it lets me just be in the moment. I'm not worried about something, whereas I normally am worried about everything. Josh Birk: I love it. All right, Kurtis, thank you so much for the great conversation and information. That was a lot of fun. Kurtis Kemple: It's so much fun. I had such a good time. Josh Birk: I want to thank you for listening, and I of course want to thank Kurt for all that great information and conversation. It was just a lot of fun. If you want to learn more about this podcast and being an admin, head on over to admin.salesforce.com. And of course, if you want to learn more about Slack and artificial intelligence, head on over to Trailhead. Thanks again everybody, and I will talk to you soon.      

Today on the Salesforce Admins Podcast, we talk to Michael Rose, Senior Director of SMB Solution Engineering at Salesforce. Join us as we chat about the ever-evolving role of the Salesforce Admin and why now’s the time to start exploring what AI can do for your org. You should subscribe for the full episode, but here are a few takeaways from our conversation with Michael Rose. The parallel between admins and solution engineers It’s always a pleasure to sit down and talk about Salesforce with Mike Rose. In his new role as Senior Director of SMB Solution Engineering, he has a lot to share with the admin community about what he’s seeing with small to medium-sized businesses coming onto the platform. Mike points out that admins and his team of solution engineers share a core responsibility: evangelism. For both, your job is to make the case for how Salesforce implementation can help your organization achieve its business goals. The integration challenges of smaller orgs As Mike likes to joke, many SMBs are running some version of what Mike jokingly calls POIM (Post-It On Monitor) integration. As in, someone comes over with a sticky note (or Excel file) and asks you to put that info into Salesforce. “That's all integration,” he says, “it is taking that data and putting it somewhere where it can be more valuable.” These workflows can be hard to change, and that’s because they work well enough to get the job done. As Mike explains, the opportunity cost of things like errors, bottlenecks, and latency doesn’t factor into the equation. It’s hard to envision a world where an entire business process could happen automatically. For Mike, the next frontier of this conversation is Agentforce. You can develop bespoke, enterprise-grade AI solutions tailored specifically for your business, but that kind of power is hard to wrap your head around when you’re still trying to limit the number of sticky notes circulating around the office. Why admins are the key to unlocking the power of AI As AI solutions continue to evolve, Salesforce Admins will play a critical role in bridging the gap between humans and technology. As Mike says, “there is always going to be a border that has customs agents and couriers and envoys working across that human intelligence and machine intelligence boundary.” Agentforce is evolving so rapidly that even the Solution Engineering team is struggling to keep up to date. So Mike recommends getting your hands dirty as soon as possible, either by spinning up a developer org or turning on Salesforce Foundations. There’s a lot more great stuff from Mike in this episode, so be sure to take a listen. And don’t forget to subscribe to the Salesforce Admins Podcast to catch us every Thursday. Podcast swag Salesforce Admins on the Trailhead Store Learn more Salesforce Foundations Setup Get Started with Developer Edition with Agentforce and Data Cloud Admin Trailblazers Group Admin Trailblazers Community Group Social Mike Rose on LinkedIn Salesforce Admins on LinkedIn Salesforce Admins on X Mike on Bluesky social Mike on Threads Mike on X Full show transcript Michael: This week on the Salesforce Admins Podcast, we're thrilled to welcome back Mike Rose, a 12-year Salesforce veteran who has a new role and some fresh insights. Mike and I get into the real talk about what it's like supporting small to medium businesses, the ever-evolving role of the Salesforce admin, and how tools like Agentforce are really changing the game. Now, whether you're a full-time admin or you're wearing five hats in your org, this episode speaks your language. Mike also shares why now is the best time to roll up your sleeves and start exploring AI in your sandbox. So you've already hit play on this episode, get ready to feel seen, and let's get Mike back on the podcast. So Mike, welcome back to the podcast. Mike Rose: It's such a pleasure, Michael. I have been a long-time listener and repeat caller, but very glad to be back after years. Michael: I know. I feel like we used to do World Tours and we'd run into each other like we need to do a podcast together. Mike Rose: It's true. Michael: And then we'd do a random podcast together. And then this last time I just saw you on Slack. I was like, I haven't had Mike on the podcast in a while. Mike Rose: It's true. And I actually said, "Yes, absolutely, let's do it." And then, I think it was a month of gap before I got back to you and then you were on vacation. By the way, folks, if you find yourselves running into the consequences of your own inaction, say, in the middle of the night on a weekend, and you say, "Oh, I really should reach out to that person, but I don't want to do it right now, because that's just awkward." Slack's scheduled messages are your friend. Michael: I'm telling you, I use Slacks' scheduled messages all the time, especially when I go on vacation, I can schedule stuff. Or the best is when you have a team that's in a different time zone. So we'll have a team, let's see when this airs, the London event will be happening/happened, and I can schedule stuff and it looks like I just schedule it right for 8:00 A.M. their time, but I'm off in sleepy land. Mike Rose: I have to say that I have one of my managers who reports up to me is on maternity leave right now, and she has set Hari Seldon Foundation style messages that pop up in Slack and they say, "Hi, it's the ghost of Gab here to tell you." And it's hilarious and it's great because it's all the reminders that otherwise I would have to remember to do, like do your Q2 check ins and do this. But what's particularly funny is that because she branded it as the ghost of Gabrielle, the reaction emoji that people are using to respond to those messages is the face of Fred from Scooby-Doo reacting in terror to a ghost. I didn't come up with this. One of her team members reacted that way. I was like, oh, no, no, no, no, this is it. Now, everybody has to react that way. It's so much fun to see these little microculture moments that build a custom or build a tradition around something as simple as an emoji response. But yes, scheduled messages, you can just sit down and knock out a quarter's worth of admin if you're going to be out and do it all pre-programmed. It's very handy. Michael: I would say, as a best practice, and I'm looking at a scheduled message that somebody had sent me over the weekend. They italicized, "Hey, by the way, this is a scheduled message." Mike Rose: Oh, that's really smart. That's really smart. Michael: Just in case you want to, not in the heat of a moment. Mike Rose: Well, it's also so awkward when there's something that you've been working on with someone and you set the scheduled message for 9:00 A.M. their time on a Monday, only to discover that at 7:30 A.M. they already gave you the full rundown of everything you possibly could need. And then, your message pops in. It's like, "Hey, just wondering if you had an update on project X." And then like, "Dude, do you not read first?" So yeah, putting it in italics is a good, I like that custom. Michael: Yeah. I do a Friday message to the team and I schedule it for 1:00 my time, and 12:55, somebody had an urgent message that they posted to the team and they said, "In the spirit of the Friday message," and I was like, oh, my Friday message is going to appear in five minutes from now and it's totally going to look like Mike forgot to send it. Mike Rose: Can't win them all. Can't win them all. Michael: Well, I'm sure Slack is something we'll talk about, but I mean, we have a lot to talk about because there's Agentforce stuff now, you have people that report to you. You've always had people that report to you, but you're demoing to different customers. Where would you like to dive in first? Because otherwise, we'll just go off the rails and end up talking about something in New York or soup, I don't know. Mike Rose: Soup would be good. Well, I could just start with on the perhaps bold assumption that previous listeners who might vaguely remember me having been here before, might think what did he used to do and what is he doing now? Michael: I'm sure people track that. Mike Rose: I actually have seen some of the Always Sunny in Philadelphia murder boards for this show, for the Admins Podcast. And they're elaborate. There's a lot of community investment in understanding the connection. So no, I've been at Salesforce 12 years, celebrated my 12th anniversary a couple months ago. Very exciting. Michael: Wow, congrats. Mike Rose: Thank you. I also am delighted to report that my brother is also now back at Salesforce. He is boomeranged, so he was part of our solution engineering world for several years. Left to go take a role at OwnData or OwnBackup. Michael: And then we acquired them. Mike Rose: And then we acquired them. So now, he is back. And so, now we have sibling force back in effect, which I'm delighted by. And so, last time I was on the program, I was managing a team of technical architects. So our pre-sales TAs are the successors of our platform solution engineering team. So these are folks who are working with customers, mostly working with the technology decision makers and technology leaders, IT leaders at customers to work on questions around security and data strategy and AI and app dev and customization and getting into the cracks or the weeds in the customer 360. A lot of conversations around integration and really with all the integration products and capabilities that we currently have and will safe harbor have in the future, the TA team has spent a lot of time working on Data Cloud and Agentforce as well. And that was what I've been doing for several years, eight years actually, I think. And then, an opportunity came along last year and I have now taken over a larger team of solution engineers, not technical architects, but core solution engineers, account solution engineers working in a part of our SMB business. So this is a team of about 50 and with several different teams, geographies and verticals within it. But basically working with customers between about 50 and 250 employees in a subset of our industry's framework to help them understand what we can offer, do discovery, do demos, talk about AI, which it seems is what we're doing a lot of the time nowadays. And it's been a great transition for me. It's been really educational and informative to get closer to the customer again with some of the opportunities and some of the challenges that are out there nowadays. Michael: And the reason, well, first of all, I love talking to you because you're incredibly smart human being. Mike Rose: Thank you. Michael: But the second is your job is so closely aligned to everything admins do. It's like an aspect of their job. You just get paid to do essentially a sales role of admin stuff. But that's essentially what a Salesforce admin does in their organization is walk around and talk with the executives and oh, well, let's see how we can use Salesforce in your area and everything you described. Mike Rose: Yeah, I mean, I think that there is a partnership between the admins and the solution engineers that's really important and that I think you highlighted very effectively, which is it's a shared responsibility of evangelism, sort of evangelism from the outside, evangelism from the inside. And the admins are those folks who are doing discovery, internal discovery, process assessment, understanding where is their pain, where is their organizational inefficiency. And I think for our small business admins, a lot of them are accidental admins. A lot of them are people who came into the Salesforce ecosystem through the back door, not really with the awareness or intention of taking on that responsibility. And so, they've had to learn very tactically. They've had to learn based on what was the problem in front of them at that moment. And when we arrive at those customers or come into those conversations, often the question is, well, not so much what should you be doing or what should you be doing better, but what are the things that haven't even been examined yet? And often, it is around something like integration. I joke with customers that we see a lot of, particularly in our healthcare customers, we see a lot of PIOM data integration models, and I'll bring that up in the course of our conversation and just let it simmer there for a minute. And sometimes someone will ask, sometimes they won't put us like, oh, I should probably explain my acronyms. PIOM stands for Post-it on monitor. And literally, the way data gets into Salesforce is someone from the sales team goes over to the lead management person, Joanne, and leaves a Post-it on Joanne's monitor saying, please update this record. That's integration. Michael: Or update this spreadsheet so that I can insert it. Mike Rose: Yes. Yeah, exactly. And so, that is an integration driving the motion of data or added value on top of data through human intervention, through re-keying, through copy and paste, through let me look over your shoulder and key this in for me. That's all integration. It is taking data that exists usually in email or Excel or in somebody's brain or on a piece of paper and putting it someplace where it can be more valuable, where it can create a multiplier effect of visibility and actionability and observability and impacting the situational awareness for business leaders and for people on the front lines. All of that is value added for that data. But when we see those points of pinch points and points of friction, usually it's because there wasn't enough perceived value to change it. It's working the way it is, why should we change it? And usually the answer is, it is working the way it is, but you don't understand or don't fully account for the opportunity cost of not having that happening automatically every time like the errors and the lag, the latency of that data and all those other things. So that's just one tiny corner of it, but it really is common, especially with smaller customers, to just see largely a great Salesforce implementation and everything working really efficiently. And then you're like, what's under this rock? Michael: Oh no, here be dragons. Mike Rose: Here be dragons. So yeah, so it's a lot of fun. Michael: Well, I think having gone through, I feel like I've kind of run somewhat of the gamut. When I started out as an admin, we had 10 licenses because that's what you got with the foundation. Mike Rose: Power of us, yeah. Michael: And then, we grew all the way up to like 1,500. So I've run, and we didn't go right away. It was steps. And so, I've been in that 50 to 250-seat area. One thing that I think we probably do, and you work at Salesforce too, we often assume that when we talk to a Salesforce admin or developer architect and they're in the smaller business, 50 to 250-seat, that that's all they do. Every day, they wake up and they pine what field can I release? What training thing can I do? But I'm going to go out on a limb and say, because you said most of them come in accidentally or through the back door. What are some of the other jobs that you find that they're also responsible for in addition to sitting in these Salesforce meetings with you? Mike Rose: Oh, gosh, that's a great question. I mean, I will say for our customers, because they're just a little bit on the larger side of the larger half of our SMB space, that by and large, by the time we get to them, they have transitioned into full-time admins. However, they're coming from sales ops or they're coming from being- Michael: They always come from sales ops. Mike Rose: They do. Michael: The sales guys are the ones that care about the CRM. That's why. Mike Rose: It's funny how that works. Sometimes it's actually an AE or someone who was on the sales team. Sometimes it's a business analyst or someone who is coming in from the marketing side or from lead gen. A substantial number of folks who end up as Salesforce admins are coming from office management, receptionist, executive assistant, presumptively, non-technical roles. But because of their proximity to the leadership and because their work tends to be a little more bursty and they have fallow time or downtime along with the time that they're extremely busy and they can do stuff asynchronously, there's a lot of folks who are coming from that front office and front of house role into a Salesforce admin role, which is great because I can't think of another enterprise software environment where it's as easy and as welcoming to get started and get skilled up than us, than Salesforce. So we do have those opportunities for people to change their careers and step in a different direction. I can think of a lot of people who were either not in IT or were in IT, peripheral IT and peripheral technology roles and Salesforce was their angle and ended up being the transformative thing that brought them into a technology career. So you'd love to see it. Michael: I mean, mine was the same way. I was in charge of inside sales and Salesforce because sales. Mike Rose: Yeah, because those are it. And honestly, if you can be effective as an admin in a part-time way, that's great because it helps you stay closer to the business and helps you stay connected to what's driving value and where the pain points are, where things need work. The thing that I think is interesting for our smaller customers is we don't have, to my mind, enough customers taking advantage of premier support, the delegated admin features and the ability to get Salesforce to do stuff for you, take stuff off your plate. I know we're adding a lot of admin AI capabilities as well to speed up the work and makes things easier for admins, but honestly, if you're going to get value out of something that Salesforce can give you, I think that's a good thing to do because what it gives you back is time. It gives you back time to again, spend more time understanding what the business needs. Michael: And there's a cost. I mean, I finally sold one of my employers on it, but the ability to, I think at the time it was log a ticket and get a dashboard made, I mean, that was an afternoon back for me. It was huge. So let's talk, I mean, we also, it's the age of AI. Everybody's got AI. We've got Agentforce. You see the headlines of big companies building their own AI. I promise you, the NAPA Auto parts in Farmington, Minnesota that uses Salesforce, I don't know if that's true or not, I'm just making it up. Mike Rose: Well, it's true now. Michael: Probably isn't building- Mike Rose: Now it's canonical since you said it on the podcast, it's true now. Michael: But I mean, probably isn't sitting down building their own AI model. What do you find are the questions or the gaps that maybe they don't know that they should look for or is there a perceived we're just not big enough so we're not going to use it? Mike Rose: It's a really interesting question. I mean, I think in general, the thing that we are telling customers or the thing that customers are coming to understand is that creating a bespoke AI solution for a particular business, for a particular sector, for a particular whatever is an enormous lift. It's an enormous lift. And I mean, again, that's true today. It may not always be true. It may not be true indefinitely into the future, probably won't be. But if you are the NAPA Auto Parts franchise in Farmington, Massachusetts, and you're saying to yourself, "What we really need is a custom NAPA Auto Parts AI," the answer is not to build it from an LLM vendor up from that layer, and certainly not to train one, again today, but you can get, it always comes down to time to value. You can get 98% of the value of a custom AI solution for your business with Agentforce today, and you get the one year, two years, three years, however long of advantage of having access to that capability, having access to the tool prior to whatever you build being ready and pressure tested and secure and highly available and supported and having someone you can call or email or text or chat for help when you need it. It's everything about developing bespoke on-premise solutions for your business multiplied by a couple orders of magnitude. Because i don't think anybody or very, very few businesses today, if you said to them, "Here's what we think you should do. We think you should start with a database solution, Microsoft SQL or Postgres or whatever, and you should build your own CRM on your own infrastructure with your own code and your own customizations from scratch." If we were to tell a customer, "We think this is the best use of your time and resources," most 99% of customers would say, "No. Why would we do that when Salesforce exists, when Salesforce is a thing?" Similarly for an ERP, I think that if you go back in time, building your own ERP or constructing your own version of that system from a database and some logic could be a thing, but then vendors came along that could solve that problem, and at the smaller segments of the market. For most business applications today, you would not assume that the right thing to do is to build it yourself even under pretty extreme conditions. Sometimes, of course you will. Stuff that is a differentiator that is going to be unique to your business, that is going to offer value that makes you something special compared to your competitors. Yeah, then you might consider building it. You might consider building it on an application platform as a service, but you might consider building it. That's where we're even further back in the chain right now with AI because the expertise is thin, because the effort is high, because the risks are very high. So you can do great stuff that's a point solution with commercially available consumer-grade AI solutions. If you're trying to integrate it and implement it in the enterprise, it's really, to my mind, only the biggest companies that have the resources and stamina to see this through and actually deliver something that is valuable, actually deliver something that's useful rather than getting something that works and does what you need it to do now and is driving value for customers now. Sales pitch over. Michael: I didn't see it as a pitch. You know what's interesting? I think of Agentforce first as a helper for admins. And listening to your answer, it was all, and Agentforce can be customer-facing. I always think of, if I could go back in time when I had 200 licenses and I had 175 questions a day, how would I create an agent that would just offload some of that lift, that I could just give it an FAQ of the last six months of questions that I've been getting that it could just answer those questions for my internal users because there's probably 10 heavy question askers and then everybody else falls in the middle. And hearing you talk through that, I was like, this is so fascinating. I wonder if I'm the only person that thinks of rolling out Agentforce first as an admin assist and then as a customer assist. Mike Rose: Oh, my gosh, yes. I honestly think that the internal use cases for Agentforce are as compelling if not more compelling than, well, I don't want to say more compelling than the customer-facing stuff. Because the customer-facing stuff, the externally-facing stuff, when it is aligned and when it is doing something that is a value add is incredible. And what we hear, see from customers is it's incredible. But it also because it's interacting with customers and it's representing your brand externally, you have to pressure test it a little bit. You have to go in and do your validation. We have great testing tools now. The Agentforce testing center is fantastic and it's a lot faster than it used to be, but you have to be reasonably confident that this is going to behave and not create brand risk. Internally, you still have that, but it's a little different. The equation is somewhat different and you can gauge the appetite and the trust of your internal stakeholders for, hey, here's something new that is intended to save you time and effort and aggravation, and we're putting it out there and we want your feedback and we want to be able to make it better. And you can do that innovation and that iteration as an admin in an agile way. You know, I mean, every admin knows who their frontier users are, their bleeding edge people, their cohort of down for anything people are. So you let them hit it, you let them work on it and try it out and see where it breaks, and then you are able to deliver that as an innovation for your organization that much more quickly. Michael: Yeah. I also, maybe this is just me, but pressure testing your AI, you roll it out and say the 500th customer, it says, "You know what? I can't find your order. Can you give me that order number again?" And they get mad and they're like, "Well, your AI doesn't know that." Think of the last time you went to anywhere where there was human interaction. I promise you, I've had to give humans behind a counter in order number more than once. And so, you think of which is worse for your brand and AI saying, "Hey, I need that order number again," or some employee that's having a bad day mouthing off to your customer on the other side of the counter. Mike Rose: Yes. Michael: We tend to forget, well, AI can never make a problem or it can never be wrong. Okay, I get that. But you're also, you got somebody, God willing, $14 an hour on the other side of the counter, what are they going to say? Mike Rose: Yeah. Well, I mean, this is funny because we said before the show we weren't going to get into the whole digital labor thing, but we're getting a little bit- Michael: No, I know. We're not. But it is one of those things where it's like you have to give yourself some sort of leeway. I've used different AIs and it's come back and I was like, but that's not what I asked. And I realized, well, I've also placed orders at restaurants. And I was like, but that's not what I asked. Mike Rose: Right. Michael: Well, humans make the same mistakes. Mike Rose: Humans make the same mistakes. But I think it's funny because there's different failure modes. There's different points where things break down or the abstraction and the expectations get a little sideways, and with humans, it's fatigue, it's long shift, it's distraction. So it's doing multiple things at once and having thing A override the logical thinking for thing B. And for an agent, it's a different kind of responsiveness and different kind of intelligence. So it's not going to get tired. It's not going to be in a different mood at 4:58 on a Friday afternoon than it is in the middle of the week. But it is going to have stronger, essentially what you might give a human support agent or human customer service rep as a boundary, a policy boundary that says you don't talk to customers about whatever, the fact that the X300 has an open warranty, there's warranty extension. The only way they find out about that is if they send it in, don't tell them in advance. And you've set that policy. That's the rule. And a human agent who's hearing from the 80th customer who says, "I've got an X300 and I don't understand why this red light is blinking," might say, "Well, you know, you didn't hear it from me. But if you were to send that in for warranty repair, even if it's out of warranty, it would get covered." Their ability to operate in the liminal space between policy and practicality is what we count on humans to be able to do. That is what we depend on human, that is what we have learned as a species over millennia is that interacting with other human beings has gray areas, it has fuzziness. And by the way, we're still not fixed as a species. We're still not back the way we need to be from 2020 and having every single interaction that we had be through a screen as opposed to in-person, because all of our heuristics are built for in-person interaction. So it's like the further you get away from talking to someone face-to-face in-person, the more opportunities there are for slippage and stuff to go the wrong direction in that gray zone. Complete tangent, I don't know if anybody saw this last week, but Google's 3D video conferencing project, project reach, project, I forget what it's called, but Google has these video conferencing booths that are hologram style 3D, like R2-D2 holograms, those are being commercially deployed, and it has already been announced, and I cannot wait that Salesforce is one of the initial pilot customers for these 3D video conferencing booths. I cannot wait. So we'll do the next podcast in one of those if- Michael: I mean, so are we talking like Star Wars where you see the person just hologram in? Mike Rose: You are look- Michael: Or maybe Marvel, which Marvel movie is that? Where they're like, there's a board of directors that are all hologrammed. Mike Rose: It's all the Avengers movies, and actually in Endgame, the Avengers themselves hologram into headquarters. But no, it's not like that. It is, from what I can tell, much more like a prison visit because well, because there's a pane of glass between you and the person. It's like you're looking through a pane of glass, but it's like they're in that space beyond the pane of glass. Michael: Gotcha. Mike Rose: So we'll see it when it arrives at New York and at HQ. Michael: Wow. Mike Rose: But my point, which I wandered away from is that human beings are perfectly equipped by evolution and by culture and custom to operate in that ambiguity. And admins know this by the way. You are always operating in, well, am I giving them what they ask for or am I giving them what they want? You have to be able to interpret and adjust on the fly. And an agent isn't going to do that in the same way. And operating in the ambiguity and operating in the gray area is where you're going to learn a lot about what does and doesn't work. So I've been thinking about this and I'll probably maybe for your listeners have a LinkedIn post on this at some point in the next week or so about what kind of careers going forward have an AI moat or have sort of a digital labor are going to be more partnership with digital labor and less being overrun or overtaken by digital labor. And I do think that Salesforce admin work is one of those categories of career where you are going to be a partner with the AI and not a- Michael: Replacement. Mike Rose: Not being replaced by it. I really think because it is that human and technology interface. It doesn't exactly matter what is happening there, but the fact that there is always going to be a line, there's always going to be a border that has customs agents and couriers and envoys working across that human intelligence and machine intelligence boundary, until and unless that boundary completely disappears, which I don't expect soon, if ever, there's always going to be work to do, there's always going to be responsibility that accrues at that boundary, and that's what Salesforce admins do. It's like you're working on that boundary. You're working on getting the fuzzy, blurry, maybe not quite explicit human stuff over to the Salesforce side. Michael: It's funny, this brings up a conversation I had with a co-worker about admins and AI and what features Agentforce is coming out with and [inaudible 00:33:06] admins and developers, because now, I mean there's AIs that write code, tons of code. I saw, I forget which one it is, you could give it just a hand drawing of an idea of a website and it'll code it for you. And some people look at that, oh, well, I'm going to be out of a job. And I think I relate things to various different industries. I remember not so much, but growing up, I grew up in the '80s and robotics and car manufacturing. That was the age. The '70s, it was still mostly men and women on a production line, putting cars together, brute force. There was some assist tools. But the '80s is really when we started to see the age of robotic welders and robotic painters and all these people are going to be out of jobs. But if you ever watch the news and they show a production line of cars, there's still hoards of people, hoards of people that show up every day to put cars together. And they're the people that know how to work with the robots. They weren't replaced with the robots. And the discussion that we had was AI is not going to take your job. The person that's going to take your job is the guy that knows or the woman that knows how do you use AI more effectively than you. It's the admin that sits down and paints the picture for the organization of how Agentforce enhances everything as opposed to the admin that sits down and doesn't understand how Agentforce can enhance anything. Mike Rose: Yeah, and I think that is, I mean, not to send people off to prompt engineering courses or getting into the weeds in Coursera or someplace like that, but I do think it's worth noting for our admins, you can spin up, I'm sure this has been mentioned repeatedly and everybody knows this, but you can spin up a developer org with Agentforce and Data Cloud in it today for free. You can have that environment. Not to put too fine a point on it, you can also call your account executive or go into Digital Wallet or My Account and turn on Salesforce Foundations, which gives you a starter pack for Agentforce and Data Cloud as well, and that's in your org and in your environment. So up to you, do you want to live dangerously and do it live or would you rather be in a nice safe developer org? But the ability to get your hands in there and start working with it is not out in the distance somewhere. It's now. Michael: Right. And a lot of that is new, thankfully. But yes, attainable. Mike Rose: Attainable. Available, attainable, achievable, and continuing to build and change and transform along with the product. And I think true confession's time, like hop into the confessional booth, it's hard for us, even those of us who are in the pre-sales world where we're supposed to be as the solution engineers, we're supposed to be as caught up as possible on what the product is doing and where we're at. And every release, we've got summer release coming out, I think next weekend for some chunk, either the first tranche or the second, it's either R0 or R1 coming soon. And I'm looking at this release in a box and thinking, oh my God, there's so much in here, this is crazy. But the thing that caught us up short as an organization was we were building Agentforce agents in the very manual topic at a time, prompt at a time mode and saying, "Well, this takes us a long time to build these demos and this is very piecework and a lot of effort to even put together one thing that is representative." And then, unbeknownst to us, the create an agent with AI button showed up in the agent builder in production. This is a real feature that's really working today, where you can, in natural language, share your website, share what you want the agent to do, and just stand back. And Salesforce and Agentforce will create a first approximation of the agent that you describe with appropriate topics, appropriate actions. Granted, it may not be wired up to your configuration, your metadata and your flows, but it's a pretty reasonable starting point and all from a website and a prompt. And that enables us to do some really interesting things. That enables us to show customers that light bulb moment and enables you as admins to go in and show your stakeholders that light bulb moment so much more quickly. Again, whether you've turned on Foundations and have the starter credits or whether you're doing it in a dev org, you can do it really, really fast. You can get to something that works the way you want really, really fast. And honestly, that took us by surprise. I did not know that that was, I thought that was still beta and still something we weren't ready to talk about, much less show. Nope, there it is in the product. So lesson learned. I feel like being an admin is a lot and being a solution engineer at Salesforce, one thing you have in common is you have to have a best served by date sticker on everything you know, on every piece of knowledge and every product awareness that you have should have a staleness timer on it, so that once this has been true for X number of years, you go back, you say, "I have to check my assumptions here. Is this truly the case?" Great example is employee cases and anybody who deployed platform licenses into their environment and was accessing the case object and had to deal with this concept of employee cases and considering how there was a lot of product specific language about that. Well, you know what? Guess what? A year ago, we got rid of that PS, the product specific language and product specific terms. And now, there's just cases. And the only rule is the rule that was there all along, which is if you as a user or as an employee are primarily or even partially servicing customers and working on customer cases, you need to be sitting on Service Cloud. Otherwise, you don't have to care quite as much of is this an employee case or is it a case case and they're actually the same object, but we're going to call them different things. No, all that is gone. And so, if you were operating with the old knowledge either as a customer or as an SE, you're going to tell people to do unnecessary work or you're going to create a lot of superstructure and a lot of effort that doesn't have to be, and that's the price of not keeping up. Michael: The price of not keeping up. That might be the title of this show. Mike Rose: There we go. Michael: Let's end on an action item. Based on your experience, admins in what we consider small, medium business size, after hearing this podcast, what should they do? Mike Rose: What should they do? Well, they should take three deep breaths and touch some grass. Oh, no, what if you're in downtown New York? You're going to have challenges. I would say first of all, they should take a moment to remember they're not alone in the chaotic circumstances, the macro factors and the micro factors that are all conspiring right now to make this a very challenging and a fraught time for a lot of people. So giving yourself that grace and space, which is what my team is focused on right now. Take a little space, give a little grace. And the other thing, the other action item is to, if you don't have Agentforce and Data Cloud active in your production environment, take a look at Foundations, see if you're comfortable turning it on. Again, remember, it doesn't cost anything to turn on Foundations, and if you don't want to do that or don't feel comfortable doing that, then by all means, please, please, please go spin up a developer org for Agentforce. It will not get easier to take the first step. It is never going to be easier than it is today. I don't know if that's true. We are in a local minima for easiness. Michael: It wasn't true yesterday. See? Mike Rose: Yeah, that's true. Michael: I mean, every single day, it's not going to get easier than today. Mike Rose: That's correct. Which I think we can look out to a point where it actually is going to get easier in that this will be on for everyone and certainly for new orgs and new environments, it'll be on for everyone. So that will make it a zero effort to get access to it. Right now, you still have to effectively opt in, but it will never get easier to decide to decide to do it. And if you decide to decide to go spin up an Agentforce developer org, you then have the ability to share and evangelize and talk confidently to your stakeholders about what Salesforce is doing with Agentforce and AI. And I think that's great. I think that's a really good place to be starting from a place of knowledge and a place of a little bit more comfort and familiarity is always going to be easier than coming at it from, well, I haven't seen it, but I understand what I've read about is, you're admins, don't read about it, go touch it, go get your hands- Michael: Or I haven't been hands on with it yet. Mike Rose: Exactly. Well, guess what? Today's the day. Michael: Yep. And you're right. I mean, there are times, I'll do the Agentforce now virtual tour stops workshops and I introduce myself and I tell people, "I've been doing Salesforce since 2006 when I remember the big release was a WYSIWYG page builder." Mike Rose: Wow. Michael: Where you could actually just drag the field from a pallet onto a canvas, and it was there and I was like, "Whoa." Anyway, those were the days. Thanks so much for being on. Mike Rose: My pleasure. It's been too long and I have had too much fun this morning doing this to let it be that long again, so I will be back sooner rather than- Michael: Okay. I will hold you to that and we will, because I should have you on before Dreamforce. I'm sure we're going to need to do a Dreamforce prep episode because you are in the weeds. You're right out there and you can help. What should we prep for? Between the two of us, we'll know what's going to go on for admins at Dreamforce. Mike Rose: I mean, I am flattered that you believe that and we'll find out if it turns out to be true, but yes, Dreamforce is going to be amazing this year. So much, so much to talk about, so much that we're going to be able to see, and honestly, if people haven't been tuning into World Tours and attending or watching, should do that because what you're getting right now in World Tour, in the keynotes and in the sessions is the Dreamforce is the run-up. Michael: The run-up. Mike Rose: It's the Dreamforce preview. Every one of those, and by the way, if you do attend and you do watch the keynotes, you do participate in sessions or the Agentforce activations or anything like that, give feedback, share it at the welcome desk, share it with the surveys that you get because that actually helps to create a better experience for everybody when we do get to Dreamforce and the big show. Michael: It's never been easier to learn than today. Mike Rose: It's never been easier to decide to take the first step. Michael: Than tomorrow. Mike Rose: Who knows? Michael: Same way. Mike Rose: It's going to be just more stuff. Michael: Same way. Mike Rose: Just more stuff. Michael: It'll be one more day. Mike Rose: Thank you so much for having me. It's been a pleasure and thanks to your listeners for being such an amazing part of the Salesforce community. We love y'all. Michael: And that's a wrap. Huge thanks to Mike Rose for reminding us that admins are basically the diplomats of digital labor navigating ambiguity like a pro. If you've ever faced the here be dragons in your org, don't worry, you're not alone. Now, here's some advice from Mike. If you can go outside, take three deep breaths and touch some grass, and then spin up an Agentforce org and test out something. Also, if you get access to the hologram booth, that would be kind of cool. Maybe try that out if you got one lying around. But in any case, make sure to share this episode with fellow Salesforce admin if you found this useful and give it a rating on iTunes. I would really appreciate that. Until next time, we'll see you in the cloud.  

Today on the Salesforce Admins Podcast, we talk to Jennifer Lee, Lead Admin Evangelist at Salesforce. Join us as we chat about what’s coming in the Summer ‘25 release and the features that will make your life easier as an admin. You should subscribe for the full episode, but here are a few takeaways from our conversation with Jennifer Lee. The Summer ‘25 release is coming soon It’s that time of year again. The time for popsicles, backyard barbecues, and the Summer ‘25 Release. So I’ve brought none other than Jennifer Lee on the pod to tell us what’s coming for admins. As always, Jen has a great blog post covering all of the changes with animated gifs that show how they work. I’d highly recommend scrolling through it as a visual companion to this episode, but the big takeaway here is that admins’ lives are about to get a whole lot easier. Jen’s highlights from the Summer ‘25 release At a high level, Summer ‘25 means fewer clicks and more control for admins. Jen highlights some key changes: The Close Case button: No need for custom buttons—you can add a Close Case button to the Case Details page and save your reps the extra clicks. Better custom object deletion: When you delete a custom object, you’ll see a detailed page listing any relationships it has to other objects. The new Permission Set Summary page: You can now update user, object, field, and custom permissions directly from a permission set’s Summary page, without navigating to multiple pages. Expanded Salesforce Go: Your guided tour for how to enable/configure features in your Salesforce edition, with resources to help you get started. As always with releases, the little things add up. And these changes help you effortlessly manage your org like never before. Powerful new features in Flow Of course, no episode with Jen would be complete without diving into the changes coming for Flow. She draws our attention to a few key enhancements for Flow: Get related records (beta): Instead of dealing with multiple Get Records and Update Records elements, you can now get entire hierarchies of related records, such as an Account and all of its Contacts and Opportunities, in a single Get Records element. Expanded resources search in Flow Builder (beta): When you enable this feature in Setup, you’ll be able to quickly find resources like fields from records and outputs from actions. New Time data type: You can now reference the new Time data type for things like scheduling reminders, routing records based on specific times of day, and triggering time-sensitive actions with pinpoint accuracy. Debug enhancements: Debugging your flows has never been easier, with element-level summaries, and search capabilities within the debug to help get your flow flowing. Approval Wizard: It’s complicated to build an approval process, so we’ve made it easier to get started with up to three approval levels, final actions, and even a recall path. There are a lot more great insights from Jen about screen flow enhancements and other changes coming in Summer ‘25, so be sure to listen to the full episode. And don’t forget to subscribe to the Salesforce Admins Podcast to catch us every Thursday. Podcast swag Salesforce Admins on the Trailhead Store Learn more Salesforce Admins Blog: Jen’s Top Summer ’25 Features for Admins Admin Trailblazers Group Admin Trailblazers Community Group r Jennifer on LinkedIn Salesforce Admins on LinkedIn Salesforce Admins on X Mike on Bluesky social Mike on Threads Mike on X Full show transcript Mike: Welcome to the Salesforce Admins Podcast. This week, Jennifer Lee returns to the pod to break down what's new in the summer '25 release. And trust me, it's more than sunshine and good vibes if you read her blog post that she put out earlier in May. From long-awaited case close buttons to a major user management upgrade, I promise you Jennifer is going to walk us through some features that make your admin life easier. Plus we dig into flow enhancements, better debugging and why thoughtful resource naming still matters. I mean, we can't not talk flow with Jennifer Lee, that's just how it works. If you've ever been wondering what to focus on with your attention to this release cycle, this is your episode. Super fun to go through. Now, if you enjoy this episode, be sure to hit that follow or subscribe button on whatever podcast platform you're listening to. So with that, let's get Jennifer back on the podcast. So Jen, welcome back to the podcast. Jennifer Lee: Always love being here. Mike: I mean, it's summer already. It feels like it was just spring. We were just talking about spring and now it's summer. Isn't this how it always works? Jennifer Lee: Well, now I'm actually feeling summer for real in Boston. It's hot out, I'm wearing shorts. Mike: Yeah, I mean, every time I've been to Boston it's either snow, really hot or snow. Jennifer Lee: And it was cold probably a few days ago too, so I'll take it. Mike: Right. Yep, that's okay. But with summer comes summer release, and we're a little behind. We're not terribly far behind. But you did put out a blog post back in May about summer '25 features for admin. So I figured we could talk through those because sometimes there's stuff that you learn after you write about it that you wanted to bring up that maybe you didn't get a chance to bring up. So that's where we're going to start. But before we do that, what other things are you working on? Anything fun you want to share? Jennifer Lee: Yeah, I'm always learning more... Like with our admins, learning more about Agentforce. So currently working on a video, should be out shortly, so we'll say that. And again, working on a new series, we should have that out not too long from now. So a lot of things in the works. Mike: Cool. I know there's always fun stuff, always fun stuff, but well, let's talk release stuff. So what were some of your top features this summer? Jennifer Lee: Okay, so for those who have case management, you always had to create your own custom close button. And it's the little things, it's the little things in these releases that make it, right. So now there's a standard close button. You don't need to go and make your own. You can just bring in the standard close button to your record page layout and boom, on your page layout, you can just say close case and it works. Again, the little tiny things. Mike: But you know what, I mean sometimes release stuff doesn't have to be like, "Look at this giant 60,000 foot tower I built." It can be, "Look at this little stool that you can now close your cases with." Jennifer Lee: Yeah, and another thing I like, so bringing me back to my admin days, is when you go and you tried to delete an object. And it just goes, [inaudible 00:03:50]- Mike: [inaudible 00:03:51]. Jennifer Lee: You got other stuff going on there, you have dependencies, you can't delete it, and then you had to go and find all the places. Mike: Yeah. Jennifer Lee: That was not fun. Mike: No. Jennifer Lee: Not fun at all. But with this release now when you try to delete an object, it'll say you can't delete it, but it'll show you all the dependencies. So all the object in the fields, and then it gives you a hyperlink so that you can go and click on that and easily find the thing that you have to delete before you can actually go ahead and delete the object. So again, it's the little tiny things like that. And Cheryl Feldman's team continues to deliver and make enhancements to user management because we all know user management can be a mess. But she's delivering stuff to make lives easier for admin so that we can focus on the cooler things that we can build out. But here are some of the highlights on the things that her team is delivering in this release. So when you're in object manager on an object, you can now go to the access page, and from there you can review and add and remove permissions from custom profiles and permission sets, and not have to navigate away. So your object access is right there, your field level access is right there. Mike: Wow. Jennifer Lee: Yeah, these are huge. Mike: Yeah, I'm looking at the GIFs that you have in that blog post. Jennifer Lee: I love making GIFs. Mike: I know it's kind of- Jennifer Lee: It really brings a feature to life. Mike: Here's a pro-tip. As you listen to this podcast, look at the blog post and it's like surround sound. Sorry, I interrupted you. Go on. Jennifer Lee: Oh, no worries. Another thing is on the permission set summary page, again, it's not having to click all these places to do the thing, but on the permission set summary, you can now update user object, field and custom permissions right there. Again, you don't need to navigate away to those places to do it. Same thing with permission set groups, on that summary page, you can add and remove permissions within your permission set group. So that's great. And then additionally on the user summary page, so on the user record, you click that summary button. From there you can edit tab access directly, and then you can add and remove assignments for the user to perm sets, groups and queues. Mike: I think that tab thing's probably the coolest thing that if you've ever had to go through and figure out- Jennifer Lee: Where it is. Mike: I mean just like, "Oh, where did I stick that tab on these 10 applications?" Or, "Who can see this?" Or why. There was always one user that just randomly had a tab show up. "Sorry, let me go fix that." Jennifer Lee: And as it relates to user management, if you haven't tried out Salesforce Go, I recommend you go and do that. That really simplifies when you're setting up your features. And it also provides access to videos and you can do tours, or it points you to some Trailhead modules and guides you through how to enable features based on your addition that you have in your org, and also whatever clouds you have. So if you have Sales Cloud, there's a Sales Cloud section, Service Cloud, Service Cloud section. So it really makes enabling features in your org much easier. Mike: So Salesforce Go, you should go check it out. Jennifer Lee: Like that. Mike: Go. Go. Jennifer Lee: All right. And no release would be complete without me talking about flow. Mike: I mean, I feel like at this point it's the sound of clouds separating, "Flow!" And angels singing. Jennifer Lee: Yeah. Mike: You're like, "Yeah, okay, let's talk about flow." Jennifer Lee: I can talk about flow all day long as you know [inaudible 00:08:10]- Mike: I know, I know. Jennifer Lee: So in the Get Records element, when you're doing Get Records for whatever object it is, you can now have the ability to get the related records as well. Right before, you would have your one Get Records for the object, and then you would do another Get Records for the related objects. Now you could do it all in one, and then when you click and you say, "Oh, I want related records," then from there there's an interface that allows you to select the fields for that record and you still train and all that. Mike: Does that apply to... Obviously it's master detail, but what about lookup relationship? Jennifer Lee: I believe it's right now- Mike: Master detail. Jennifer Lee: [inaudible 00:08:57] detail. Mike: Okay, that's what I figured, but still... Jennifer Lee: Yeah, it's still big. Mike: Get Records account. Hello. Jennifer Lee: Yep. So when you are finding your resources, let's say in your flow you have a lot of things you want to look up or related and all that. You now can use the expanded resources search, which you have to enable and set up, and then it'll be available in your flow to quickly find resources. So one thing to note was we did push this out, so something similar to this in winter and we took it back, and then based on customer feedback, we brought it back and improved it. Mike: Do you have... While we're talking resources, do you have... Because I was looking at your GIF, and I would imagine you have a methodology for how you name your resources. How would you advise admins name their resources? Because obviously this is super powerful. Jennifer Lee: Yeah, so when I create my own custom resources, I do have a certain name in convention, so if I have variables, I'll do a VAR in the front of it, if I can easily find those. Or if I have a record collection, I'll always have the word collection in there, so I can easily find it. Or if you're using a text template, the word text template or a formula. So I'll just put those little keywords so that when you're doing the search and you know it's the thing that you're looking for, you can easily type that in and then they'll show up in the search. Mike: Yeah, no, it's smart. I like that. Thank you. Jennifer Lee: And then for people who have flows that deal with time, we now have the time data type that you can use, and you can also reference time formulas as well in your flows. So that's huge for people who have those time dependent flows. Mike: Wow. Jennifer Lee: Yeah. Mike: Oh, that gives me a little bit of anxiety. I see minute is an option. And second, holy cow, second. Bless you people that have time dependent flows. Jennifer Lee: Yeah, I had a few of those as a customer. Mike: Okay. What would be an example of that? Like was it on cases mostly? Jennifer Lee: Yeah. Mike: Because you need to follow up or it needs to do something like hippity hop right now. Jennifer Lee: Yeah, we were a big Service Cloud customer, so yeah. Mike: So based on how old that case is or how long that's been, yeah, I could see that. Boy! Testing. Jennifer Lee: So the flow team has also made enhancements to debug. So now it has slightly different UI, but it's so much easier to debug your flows. They now have element level summaries, and you also have search capabilities within the debug. So let's say you're working through and you have a variable that you're just checking to see if the data flows through and the various elements. You can do a search on that variable name and it'll pop up and kind of highlight where it is in the debug elements so that you can go through. You can also show API names and things like that, so it's really more interactive than it was previously. Mike: Yeah, that makes sense. Boy, that's got to be helpful. Jennifer Lee: And for admins who have approvals as part of your process, there's now a new wizard, will help you build your flow approvals. I never really use approvals, like I never really had a need to use approvals in my business, but the process of going through the approval process seemed really clunky. Mike: I built approvals a few times, and the problem I always had is the company kept changing the approval process and not telling me. So that, or making the approval process something wonky, that was super hard to build. Jennifer Lee: Yeah, I would believe that. Mike: You're like, "Yep, been there." "Did you even consult me?" "No, we just changed this." And so now... And then you always had users that were like, "Well, but that doesn't go to the right person." I know, they changed that without telling me. Jennifer Lee: And I was supposed to know. Mike: My time machine... What was that? What's that... I was thinking of that. Cerebro from X-Men. "My Cerebro is in the shop." Jennifer Lee: All right, so a couple of screen flow enhancements. So you now have the ability to trigger a screen action without needing a button. So previously we had a button that was in beta, so now that's generally available. So go ahead and use it on all the places everyone. There's also now a new feature called File Upload Enhanced (Beta) component. It's in beta. You can bring in the component and with File Upload, you can now set it to require files to be uploaded. So if you have cases and you have a screen flow, you want people to upload files, previously they could forget about it. And then you're like, "Oh, well I wanted you to give me supporting documentation," but there's no way to require that, and now you can. So note that you do need to enable that feature in Salesforce Files and set up, and then it'll show up in your screen flows. Mike: So then that puts the file on the record? Jennifer Lee: Yes. Mike: Oh, man. These are all things I needed 20 years- Jennifer Lee: Back then, right? Back then when we were admins. Mike: Oh, I so... Oh, just the number of times I would have... Well first of all, we did have screen flow. I envisioned what a screen flow would look like, and then they'd be like... Then at this point, it should be like a website and you upload a thing that maps out all the sites. And I'd be like, "That sounds really cool, I wish I could build that." Now it's 2025, now I could build that. Jennifer Lee: Yes, without a developer. Mike: And the file just goes to the record and away we go. Jennifer Lee: Yep, yep. And then lastly, on screen flows, you could... Like imagine your screen flows, you have multiple columns and fields. Let's say you're collecting an address, the street, zip code, state, those are different sizes. You typically don't have a long field for a two-state drop down [inaudible 00:16:20]. So now you have the ability to adjust the components and the fields on your screen flows to say, "Here's the width I want to use," and then do the vertical alignment too. So it really makes your screen flows look so much prettier- Mike: Oh my god. Jennifer Lee: ... than before. Mike: And just useful. I'm watching your GIF, and the number of street, city, state zip fields that I've had in orgs that are all the same size. And everybody's always like, "Well, how come you can't make it look like an address field?" Well, you see there are reasons, but not anymore. Jennifer Lee: Yeah, that's my wrap up. Mike: Those are cool. Jennifer Lee: But there's more in the blog. Mike: I mean there is, you scrolled through a lot. There's always a ton, and it's a ton... And it's always based on like, I think it's fun when you go through the stuff, the release notes every year and you kind of look at it. Because the stuff you always get the most excited about are the things that frustrated you that you kind of knew weren't hard to fix. It was just like at some point you know it's going to bubble up on the calendar and somebody's going to knock it out. I think that styling for flows is huge. I mean flows for the... Screen flows for me are the easiest way I can get into flow and kind of understand it because you can visually see what you're putting in, but the ability to edit the way that looks has always been not the most easy. But this now... I like your example, you're like, this is a really long label for a long text area. Yeah, I mean, it spells it out, but it's really helpful to understand because the end user's the one that's going to stare at it all day. You're going to look at it a few times in testing, and you also talk about, "Oh, icons. Oh, these are cool too." Jennifer Lee: Oh, those icons that you can use... So if you have a pick list in your screen flow, you can add little icons to make it easier as well. Mike: It's funny, I was on a related unrelated topic. I was at a car show a couple weekends ago, and I ran into somebody and we were talking. He said, "Oh, well, I code for the university because I live in Iowa City." And I said, "Oh, cool." He asked me, "Where do you work?" I said I work at Salesforce. And so immediately he thinks I understand code. I wasn't going to say otherwise. But the next words out of his mouth really caught me off guard. He said, "You know, your lightning design system is probably one of the most well-thought-out documented systems I have ever had to work with." And I bring that up because the images of those like, excellent, good, like the way that our iconography kind of looks within the system, I think is really cool. So is a long walk for a short drink of water, but... And it would carry over. Plus anytime that you can mix things up and use images... I mean images, do you remember back in the day, you used to set up a resource library and import all these images, like four star images and five star images and stuff, and then you would show that on the screen. I used to have users that were like, "You're a wizard. I can now see stars on my account page." And I was like, "Yes, I am." Jennifer Lee: I do remember that back in the day. Mike: Because the edit page would just show them a number, but the actual page... Anyway, off on tangents. Off on tangents. Jennifer Lee: Oh, I do have a note. So there's one thing that is in summer that I didn't include in my blog because I wrote my blog and then I found out that it's in this release, but Prompt Builder undergoes a major UI change. So for those who use prompt templates, check it out. It is a better UI right now in summer. So think of the little preview section that you have, the resolution that has its own tab now, and you don't have that little window that you have to scroll to see everything. Yes. Mike: Okay. I like that. I was going to say it wasn't that bad. And then I was thinking, "I've done a bunch of those Agentforce NOW tours. Jennifer Lee: Yeah, like this itty bitty thing- Mike: Well, there's always a part... There's kind of two parts on that. The intro of the Agentforce NOW tour that I do. One of them is shortly after you enable the agent and then you have to show the drop-down open agent and builder. And it's literally right at the bottom bar of your web browser. So I would always tell people like, "Look at me in the camera," and I would point and be like, "It's at the very bottom because you're probably looking at a screen and you can't find my cursor. It's at the very bottom." And then there was always that. Then the second time was when we would test prompts after we've built our first prompt and we brought in some fields in that resolution window, I'd always point on camera down at the bottom. Like, "It's down at the bottom all the way on the far right." So sticking those in their own... You said their own window? Jennifer Lee: They're like little plus and minuses, and then they'll expand out to show the [inaudible 00:22:42]- Mike: So much nicer. So much nicer. Jennifer Lee: So the preview button has also moved up to the upper left-hand corner. Mike: Okay. Yeah, because where it was before kind of didn't make sense. Jennifer Lee: So you can hit this little icon and then set your record and then you'll hit the preview button next to it, and then it'll do all its magic. Yeah, folks should definitely check that out because it's a great improvement to what it is today. Mike: Absolutely. Jen, thanks for coming by and pointing out some of the newest features. I mean, I'd read your post, but it's kind of always nice to hear from the person and you know, like, "Oh, here's what's really cool," because sometimes texts on a page just don't convey it. Jennifer Lee: Yeah, and I just like geeking out about Salesforce so I can talk about- Mike: I know. Jennifer Lee: ... [inaudible 00:23:44] all day long. Mike: I think back to the days of going through the release notes, and now I have this fun competition, and when I say fun competition, it's fun for me. I like to go through the release notes and then I like to feed the release notes to an AI, and be like, "What do you think are the top features I would like or top features for admins?" And I like to see what it finds versus what I find, kind of like a treasure hunt. Jennifer Lee: And what did it do for this release? Mike: It was kind of a little all over the place. Yeah, it wasn't as concise as yours. It also doesn't understand how cool flows are, and so it points things out. I will say this, it's very helpful in getting through sometimes a lot of release notes and finding things you might've not read. I think that one of the coolest pieces of advice I was ever given by somebody was, read all of the articles front end to end that are on the front page of a newspaper. Because most of those articles, very few of them are encompassed all on the front page, but they get you into the newspaper. It's also a good mixture of topics, so it might be things that you not normally read, and I think it was good advice. That is kind of like... It's like the AI version of it. Sometimes you got to feed the release notes to some AI and just see what it points out because it might be things that you scrolled past and had no idea. Jennifer Lee: Because there's a lot to go through. Mike: I mean, the other part of it too is it's an enterprise level platform, so you kind of want it to be a lot of release notes. I do anyway. Anyway, thanks for coming by, Jen. We'll be sure to check out your videos on YouTube because they're always super explanatory, and they're helping people understand agents and flow, and how agents use flow and how you can put flow into agents. Jennifer Lee: Well, thanks for having me, Mike. Mike: You bet. Big thanks to Jennifer for walking us through the summer '25 updates. It's always a treat to hear her perspective, and it gives that blog post so much more life, right? There's a lot of things that we find out. She walked us through fine-tuning flows and trimming down clicks and user management, and just discovering the magic of the time data type. This release really has something for everybody. You can hear me go through my brain on writing flows with that time data type. That was... I had a moment, I will just say. Be sure to check out Jen's YouTube videos because she dives through so much stuff and they are so incredibly useful, and I'll include a link to her blog post as well. And of course, if you enjoyed today's episode, go ahead and give it a share. Maybe send it to your friends, your admin friends, or other people in the community that would benefit from being up-to-date on the summer '25 release features. So with that, until next time, we'll see you in the cloud.  

Today on the Salesforce Admins Podcast, we talk to Joshua Birk, Admin Evangelist at Salesforce.   Join us as we chat about why your metadata is crucial for building effective AI agents.   You should subscribe for the full episode, but here are a few takeaways from our conversation with Joshua Birk. Why multi-tenancy still matters for Agentforce As the Admin Evangelist team has been helping people get started with Agentforce, we’ve noticed that the key to unlocking this new technology is to revisit some of the oldest concepts about the Salesforce platform.   That’s why I brought Josh Birk on the pod to talk about metadata and multi-tenant architecture. If you need a refresher, that’s the idea that Salesforce is like an apartment building where each org is an apartment. Your stuff is in your individual unit, but the entire building shares resources like water and electricity.   So what’s the difference from 2010? As Josh explains, it’s that every apartment comes standard with an Agentforce-powered robot butler. Quality data leads to better automation Imagine you’re sitting down for dinner, and you want your robot butler to set the table—how does it know where the forks are? And what happens if they’re buried in your junk drawer?   Clearly, a robot butler will be more helpful if you keep your apartment organized. And, as Josh points out, the same is true for your Salesforce org. AI agents rely on your metadata, like description fields and field types, to help them respond correctly and find what your users are looking for.   With longstanding orgs, there can be an “if it ain’t broke, don’t fix it” mindset, but that’s the equivalent of throwing everything in the junk drawer. Doing a little spring cleaning and organizing your metadata helps Agentforce help you. Why you’re already an AI builder The key thing Josh wants you to realize is that you’re already an AI builder. An agent is just another user in your org, and so the work you do to make your data easy to use is also what powers the solutions you build in Agentforce.   That’s why it’s so important to fall back on Salesforce fundamentals. Building an agent is the easy part. The hard part is making sure your metadata is in a good place to support your AI solutions, but that’s the work that admins do every day.   There’s so much more great stuff from Josh in this episode, so be sure to take a listen. And don’t forget to subscribe to the Salesforce Admins Podcast to catch us every Thursday.   Podcast swag Salesforce Admins on the Trailhead Store Learn more Trailhead: Custom Metadata Types Basics   Admin Trailblazers Group Admin Trailblazers Community Group Social Josh on LinkedIn Salesforce Admins on LinkedIn Salesforce Admins on X Mike on Bluesky social Mike on Threads Mike on X Full Transcript Mike: Hey, Salesforce admins. Ever wonder what multi-tenancy AI and your junk drawer have in common? Luckily, Josh Birk is back to explain it all, from forks, to metadata. Yeah, we even throw in some robot butlers. This episode's a ride through the architecture that makes Salesforce magic happen with, I promise you, enough analogies to stock your kitchen. So, if you've ever said, "Wait, where are the forks?" This one's for you. And when you listen to it, that sentence will make sense. So give it a listen, send it to your friends. Be sure to hit that follow subscribe button to get brand new episodes downloaded on your mobile device. And without waiting any longer, let's get Josh Birk back on the podcast. So Josh, welcome back to the podcast. Josh Birk: Thanks, Mike. Glad you're back. Mike: It's been a while, but you've been working on stuff. Josh Birk: I have been working on stuff. It's been a busy little quarter. This thing called AI never really stops sleeping. I guess it's one of its benefits. But yeah, trying to catch up with all things AI, and data cloud, and especially trying where there's a wealth of stuff happening before Dreamforce, and we really would like to get our admins community armed with that information. Mike: You mentioned Dreamforce. Dreamforce start till October. Josh Birk: Well, I know, but I thought July was a really far away away, and I realized I have a trip to Montreal next week because it's June, and it's like, "Okay, right." The months, they're collapsing away. Mike: It literally, it's like one minute you're like, "Yay, it's February," and the next thing you know it's like 4th July. Josh Birk: Right. Yeah. And you have a TDX going over. Mike: All the hangovers. The first thing that I think we want to talk about, so what's crazy is we brought this stuff up, what are we going to talk about, and you're like, "Let's talk about the multi-tenant analogy." And I was thinking back to, whoa, that was like 2010 when I first learned about multi-tenant, and hearing the Salesforce apartment analogy. Josh Birk: Yeah. Yeah. It's interesting. It's actually, I think, a testament to the platform, that you can use it for years and not really understand. And I used to understand not having to comprehend exactly what's going on, because one of the, it's not really a catch 22, it's this nice cycle because it's part of the wonder of the multi-tenant structure is that you don't have to worry about it, because you log in and you're in, basically, your own space. First of all, I think it's one of those things that does get bandied about, but sometimes it doesn't stick in people's heads. For instance, I was talking to somebody about Trailhead, and they were like, "Well, is that why we have multi-tenancy?" And I'm like, "No, no, no, no, no. Multi-tenancy was the thing, that was day one. That was the magic fuel that was moving our product before anything else was getting built on top of it." And I think it's important to kind of have a basic understanding, because every now and then, as a developer, one of the things that people, if you come from Java, or you come from C++, and one of the things you're used to doing is whatever you want, if you want to spin up a crazy thread in an application and have it run for three days straight, the only person stopping you from doing that is you. And so then people come into Salesforce platform and they're like, "Well, Apex has these limits to it. You can't spin up too many CPU cycles, you can't do too many searches." You can't do these things. And some developers are like, "Why are you tying my hands back behind my back?" And it's like, because part of the multi-tenant architecture makes sure that Mike's org might be sitting next to my org in terms of hardware. So we're sharing a CPU. And if Mike misbehaves too much, you might steal CPU cycles from me, and vice versa. You might slow down my application by no intent of yours. You don't even know I'm on the other side of the wall, but you're using up all of my power. And our multi-tenant architecture basically makes sure that doesn't happen. And it makes sure that you can go in and do all the things that we give you the rules to, and I can go and do all the things we get the rules to, and we're always going to be play safe with each other. Mike: And I remember, it's so fascinating here you describe multi-tenant that way, because I remember having to describe it as, "No, no, no, no, all of the data that we put in is in our own apartment. And all of the data that somebody other company puts in is their own apartment. And it doesn't matter what pictures we hang on our walls, they can't see our pictures and we can't see theirs." Josh Birk: Exactly. Mike: As opposed to the amount of boiling water that we have left, because we're all sharing a water boiler. Josh Birk: Right. Because it's summer in Chicago, and you take three showers a day sometimes. Yeah. Mike: Right. Josh Birk: Yeah, exactly. Mike: But I mean, perspective wise, I'm thinking through all this, and multi-tenancy, if you're going through even Salesforce admin one-on-one stuff, like this is the first day, everybody's thinking AI and agentic, and what can AI do for me. What are you doing talking multi-tenancy when we have this really cool agentic stuff that we should be talking about? Josh Birk: Yeah. And it's a great question, and there's a couple of answers. And the first one is to kind of go back to the analogy of an apartment complex. So that experience starts at the front door. You can't get in through the front door unless you are a tenant in the apartment building. And then the next thing you're going to do is you're going to go to your apartment, and you can't get into your apartment unless you're a tenant of that apartment. And then start breaking all it down to the stuff that you were just saying. Your pictures are your own, the boxes you have in the attic, they're all your own. All of that is, so this is why trust is our number one value. You trust us to hold your apartment near and dear to you, and you trust us to keep that security layer, you trust us to keep the powers on. All of that is the same stuff that the agents are building on top of. That is why they're a very easy to use enterprise solution for AI, because they're like the robot butler who lives in your apartment. And all of that other stuff, the security, the power, the storage, all of that stuff, it just comes with the apartment. And your agent is able to understand where all your pictures are, and where all your boxes are, and how those boxes are structured. And one analogy I like to give is like, well, if you're going to ask your robot butler, aka agent force, for a fork, if I'm a guest in your house, like, "Hey, Mike, can I get a fork?" You would know which drawer to tell me. You would be like, "Go to the second drawer next to the stove. That's where the forks are." How does the agent know that? And the agent knows that because of metadata. And metadata is the second tier of what makes multi-tenancy work. One of the things back in my workshop days, a question got a lot, especially because I'm talking to old school Oracle developers and Java developers, and they're like, "Well, what's your web stack?" What they're asking is, are you running Java? Are you running Oracle? And stuff like that. I'm like, "I could answer this for you, but it's not going to be the answer you think it is, because our data structure doesn't start like a normal database does. Our database structure starts with multi-tenancy with a metadata tier on top of it." And what that metadata allows you to do is you can put your forks in a drawer, I can put my forks in a different drawer, and metadata can tell the agent, "This is where those forks are held." And let's say you ripped your kitchen apart. Now, if you're going to rip your kitchen apart and put your kitchen back together again, you're going to get a contractor. You're not just going to go pay a couple of teenagers with some sledgehammers and just knock everything down. I mean, it might start that way, it could be entertaining. Mike: I mean, if you watch HGTV, that's how it's supposed to go. Josh Birk: It does kind of start that way, right. But at some point, you're going to want to have a plan for how tall the counters are, where the counters are going to stop, all of this kind of stuff. You're going to have a blueprint. And a blueprint is going to instruct everything down to where your drawers are. And that blueprint is crafted for you in metadata whenever you're doing things like creating custom objects, and you're creating custom fields. And everything on the platform basically has this blueprint to it. Now, the great thing about that blueprint is if you invite me over to your apartment and I'm like, "Dude, this is a killer kitchen. I would love to have this kitchen in my apartment," you're like, "Dude, I'll just get you the blueprint." And then we just package up that blueprint, I take it over my apartment, and then I can just build out the kitchen exactly to your scale. Mike: So, I'm thinking through all of the times that I've built and remodeled my kitchen, as admins do, continuing the metaphor, and failed to write blueprints. So now I'm building an agent and I'm telling it forks are in the second drawer from the left of the stove. And that's the instruction I give it, but it has to go learn what the second drawer is because it doesn't have a blueprint to look at. Josh Birk: Yeah. So this is a very strange little section of metadata and AI. It was one of the first things that I heard during a workshop, and it's always stuck with me. So you've got your blueprint. Your blueprint knows the size and scale of your kitchen, it knows how many drawers you are, and it knows where the drawers are. Now, let's say you want to identify which drawer is going to be the silverware drawer. When you're building out the custom object and you're building out the custom fields, we know that it's good to have a description fields. So if there's any ambiguity, because sometimes you might have a similarly named field, you might have a similarly named object, something like that, and you want to make sure people are guided to the right drawer, so find that fork, well, if it's a good idea for a human, it's a great idea for an agent, because when you say, "Get me the largest fork in the drawer," it's going to go through what it knows about the data model. And if it sees the word fork in a description, it knows it's going in the right direction. You're giving it that breadcrumbs. And this is why it's like, "Which descriptions fields do you mean, Josh?" And it's all of them. It's the ones in the custom actions, and we won't even get into instructions yet. It's all the ones in your custom fields, it's all the ones in your custom objects, et cetera, et cetera, et cetera. You want to leave that little breadcrumb trail so that when agent force is doing its little agentic reasoning, and it's like, "Which custom action should I use? Which standard action should I use?" You're going to end up with a much higher rate of success. Mike: So what I'm hearing is the importance of metadata in terms of describing things, in any possible term that a user could use to give that to an agent, so that an agent doesn't have to reason, "Oh, date on opportunity means contract date, as opposed to close date, because you have six date fields and you title them date one, date two, date three." Josh Birk: Exactly. Mike: And the user, because it's nomenclature within the organization, oh, well, the second date is the date it leaves the warehouse, and the third date is the date it's expected to be delivered. But if you ask an agent that it's not going to know that until you tell it through metadata. Josh Birk: Until you tell it through metadata. At best, if it doesn't know, it will guess, which is problematic for two reasons. One, obviously it might be wrong, but it also might be right two times out of three, but then you don't know that third time. It makes your testing even harder anytime that the agent has to guess for it. Now, people might be listening to this like, "Well, okay, I understand that, but how powerful can a description field be?" And it's true. It's only so powerful, but the great thing is, this is why, and it's one of the things I think people, I really like talking about this, because one of the things is I think it demystifies agents a certain extent. We treat agents almost like, I just referred to it as a robot butler. So we're already into a sci-fi. We've gone from an apartment complex to Star Wars, right? Mike: I'm thinking Rosey from the Jetsons. Josh Birk: Yes, yes. I love it. I love it. But there's so much human in the loop that's involved. And we have to remember when we talk about guardrails, all agents have guardrails. It's the thing that keeps them from saying horrible things and doing horrible things. And there's so much we can control over an agent just by text, just by writing to it. And so we've been playing with agent force internally. And one of the first things I had to do, because I was trying to make sure that our users could talk to it like they would want to talk to it, right? So when you say my blog post, you mean a blog post that our wonderful Eliza Riley created an object record for, and then added you as an author, and then added Kate as a reviewer. So we're really talking about contributors. But if you say the word my without giving any context to the agent, it's going to think you mean the current owner. But instructions, you can say, "Hey, in this topic, when I say my, when I say blog, I mean these different things." And so that's the first place that the breadcrumb trail is going to start going, and it can be one of the little behavioral issues. It's not responding in the right way, or it's not using the right style and things like that. The prompt builder instructions can solve so much just by writing a few sentences. Mike: So, I'm thinking through the apartment analogy. And I've used Rosey before, because I do think eventually at some point we'll have, I would like to have a robot in my house. Amazon had that little puppy dog thing. Josh Birk: Yeah. Mike: I should have signed up for that. But in hindsight, I don't think AI was really where it should be. Josh Birk: It was basically an Alexa on wheels, I think. Mike: I know. I was afraid it was going to go downstairs and there goes $2,000. What was that, literally here's the sound, it was like, "Zoop, zoop, pom, pom, pom, pom, pom, pom."And then mom would be like, "What was that?" And I'd be like, "That's the sound of $2,000 falling downstairs." Josh Birk: Not being able to use stairs. Mike: You can't use stairs. But I would like to have that. So, I'm thinking through the apartment analogy. And basically what you're saying is, so Salesforce can roll out the ability to find forks, to all of its agents. And so everybody in the apartment complex benefits, but what's key is your blueprint, because where you keep your forks is different than where your neighbor in 5B keeps their forks. And so if you are filling in your metadata and Salesforce ups the ability for agents to do things like find forks, like, "Oh, now they have the ability to find spoons," everybody can benefit all from one action. But just because of that action doesn't mean that your agent is up and running because it needs the metadata to really target in on what does this action actually mean for me and the customization I've done. Josh Birk: Right. Mike: Is that fair? Josh Birk: That's fair. And remember, your Rosey and my Rosey, they're the same model. And if there's upgrades, you're going to get your Rosey upgraded, I'm going to get my Rosey upgraded. Mike: Kind of like our phones. Josh Birk: Like our phones, right? Every now and then, it's like, "Oh, yep, you're three updates behind. Please restart your phone." Mike: Any more it seems that way, doesn't it? Josh Birk: It really does. They're coming faster. Mike: I never used to fall behind. The other day I was like, "Oh, you're three updates behind," it was something like that. I was like, "How?" Josh Birk: "I thought you were doing this for me." But even if I copied your kitchen one to one, but we also then give me the ability to swap out the drawer that forks are going to be in, or to add a new cabinet and things like that. And the great thing is about metadata is my Rosey instantly knows my new kitchen layout because of the metadata. Now, going back to the blueprint and the description fields and the topics, and also the other thing to mention here is the quality of your data. How well-structured is your kitchen when it comes to, do you have all the forks in one drawer, and the spoons in another drawer? Do you have one of those nice slotted trays? Mike: What do you do if you have a junk drawer? Josh Birk: What do you do if you have a junk drawer? And it's like, what if you have everything in the junk drawer? Well, Rosey's going to have a really hard time finding your fork if everything's in the junk drawer. So, all of these things we're talking about, metadata, description fields, quality of instructions, guard rails, we also have to all go all the way back to the OG. Quality of your data structure, right? It's great to have a description field to help them guide, but it's also a really good idea. Do you need three fields that end with one, two, three, and four? Or is it now the time to convert that into a pick list value, which might be easier for Rosey to work with? Mike: Yeah. I'm just thinking through that, because I was also thinking of how many I got stuck on junk drawers. Josh Birk: We're Midwesterners. Is it a Midwestern thing? Mike: You ever ask a question and then somebody totally answers, and you're like, "Sorry, I wasn't listening. I was thinking of a junk drawer." Josh Birk: Well, see, and that's now I have to fall back on my old, "Do we have to explain what a junk drawer is? Do people know what a junk drawer is?" Mike: I mean, it's like everything that doesn't make sense. My junk drawer is the water filter for my fridge, my owner's manuals, a pile. And I mean a pile of pens and pencils from various local establishments. And they're in varying states of operability. So, that's why you have three, or I got at least 10. 10 pens. Probably three of them work. And the other seven are just there, it's like a lottery. If you get it and you go to try and write something, and that's how it gets thrown away. And then straws. Straws, and then packets, ketchup packets. Josh Birk: Ketchup packets. Mike: Hot sauce packets. Josh Birk: Hot sauce packets, hot mustard packets, anything from a Chinese store. Mike: Right. And then here's where you throw in, it's a real wild card, this is where the podcast goes completely off the rails. You throw in the occasional thumb tack rubber band. Josh Birk: Just put traps in for yourself. Mike: And the newest thing now is those 3M wall sticky hanger things. Josh Birk: Oh, yeah. Those have escaped. Mike: That's a junk drawer in the Midwest. Oh, and a flashlight. Got to have a flashlight. Josh Birk: A flashlight. And a flashlight. Mike: Flashlight that may or may not work and may or may not have batteries in it. Does it matter? Still in there. Josh Birk: Still in there. Still in there. Yeah, that's about it. Mike: But no, so back to the regular scheduled podcast, already in session, I was thinking through the junk drawer of like, the junk drawer to me in any database, or mostly in all of the Salesforce databases I've managed was the description field. People throw everything in the description field. And then I can foresee now, God, the nightmare of, "Well, I asked Agent Force, what was the last five calls that I made, and it couldn't find it." "Right. Well, did you log those as activities?" "No, I just added it as bullet points in my description." Josh Birk: Exactly. Yeah, and it's like, we're humans. And the amount of interesting human behavior I've seen over the years of how people have implemented Salesforce. But there's kind of the old adage, if it ain't broke, don't fix it. And if I'm doing this day in and day out and it's working for me, great. But the times they are changing. And this is back when we were trying to get people to move from visual force to web components, now aura. I'm like, "Don't fear it. I'm not asking you to go and delete all of your existing components and start from scratch and build everything up with all of these amazing demos that I'm talking about, but now is the time to start getting your feet wet. Now is the time to start looking ahead. Now is the time to go to Trailhead and tinker with your first agent." "And now is the time to start asking really hard questions like, how messy is our data structure? And what have people started?" So, I'll go way off on a tangent, but I swear I'm getting somewhere. Back when I worked for a major retailer as their lead client side developer, we had a problem with pop-up windows. And the problem was, because everybody knows- Mike: Got pop-ups for a long time, were just the bane of everybody's existence. Josh Birk: Right, and they get hidden behind other browsers, they just do all this behavior you don't expect. Mike: Pop-unders. There was pop-unders too. Josh Birk: Yeah. So me and my old friend and colleague, Jamie Dehansen, he was the designer. And so we started a project to take that down and turn it into a modal window that would be on the page itself, that Jamie and I could control 100%. The event structure, everything clicks. We launched that thing, we're so proud of ourselves. This thing is cool, it looks great, it works wonderful. And then the complaints. Started swarming in. I don't mean two, I mean like 20, I mean like 30, just over and over again. Because people kept asking, "Why did you take your comparison feature away?" And Jamie and I are like, "What comparison feature? We don't have a comparison feature." And then we realized that people were going to a product item, clicking the pop-up, going to another product item, clicking the pop-up, taking the two pop-ups, putting them next to each other. They had built their own feature without even asking us. And so we fixed it by adding that functionality in, like the way we wanted it to work, and everybody was really, really happy. But that's the kind of thing that happens within a Salesforce org. You didn't think that you gave them the feature of transcribing their phone calls into the long text field that was meant for something else, but you did, and they're using it. And that's the kind of, I think you're credited for saying the walk around the broom style of admin. Mike: Sabla. Josh Birk: Sabla. You need to look over somebody's shoulder and be like, "That's not how I meant you." We need to fix that description, that long text field, and I need to make it work in our data structure so that Rosey can find those calls for you. Mike: So, here's a fun fact, and this is a deep cut for people that listen. So, I came up with Sabla after watching planes, trains, and automobiles. Josh Birk: Oh, really? Mike: Yeah. Let me tell you how. How does planes, trains and automobiles open? Opens with an ad agency in downtown Chicago. I think, right? Josh Birk: It's New York. Mike: It's New York, okay. New York. The CEO is looking at what would be a billboard, except he's sitting at a chair in a well-lit office room. My friend and I were watching this, because we watch planes, trains, and automobiles every Thanksgiving. And he owns an ad agency, and he just dies laughing every time he sees that, he goes, "So what's funny about this is how executives look at billboards, but everyone else is going to look at this billboard going 60 miles an hour down an interstate trying to make a corner and avoid a semi." He said, "Because nobody looks at billboards from the perspective of the person they want to sell to." And I was like, "Oh, it just dawned on me," because I had just done a Salesforce deployment. I was like, "I didn't look at that page that I deployed. I looked at it from my desk with my headphones on, uninterrupted, knowing what I was looking for, not on a phone call with an irate customer because their plant had just died or something, and trying to work through how do I disseminate this information." So there you go. There's the history of Sabla. Josh Birk: Love it. Love it. Yeah, the interview I did with Katie Coats, she was like, "Don't tell me how you've used this application. Just show it to me." Mike: Right. Yeah, yeah. Okay. So put a lid on this. Josh Birk: Yeah. Mike: All the admins out there are sitting like, "I was supposed to deploy Agent Force three weeks ago," or, "We're still trying to figure out how we're going to Agent Force, but my CEO wants it deployed by Dream Force." Josh Birk: Yeah. Mike: What should I be doing now? Because it feels like I should be learning to build agents, and you should. Josh Birk: You should. But, we go to these workshops and we're like, "You're already an AI builder. You're already an agent builder." And the reason I can say that confidently, and this is why we're talking about things as basic as your data structure in your description fields, remember your training, all of these things that you already do when it comes to who can come into the apartment? Who's allowed in the apartment during certain hours? Who can go to the certain rooms? Who can use the certain? All of these things is the stuff that the agent is going to build up on. So the better you're already doing your job, the better the agent's already going to be. You're already putting in the hard work, that's the hard work. Building the agent's easier. Way easier. Like Jen and I have compared notes, and most of the flows we work with agents are fairly intermediate in complexity at bust. I've got a few that are five or six steps at all. So, if you don't even have to be Jen League, Queen of Flow, you don't have to be a crazy flow-natic. You just need to know how to build flow. Now, you know how to build flow, great: you know how to build custom actions. And when you take that flow and you convert it over to a custom action, all of that work that you've put into the apartment leading up to that moment is going to make that agent so much more successful. So it's like, don't think of it as this brand new thing that's off on its own little lonely island, it's more similar to when we released Flow, or when we released things that are new, but they're new on top of the platform. And so all of the things you already know how to do on the platform, all of those things are going to be important. In one of the blog posts we have coming out, one of the things Jen reminded me to put in there is, "Remember to think of your agent like a user. It's a Rosey. It's another thing in your apartment running around doing stuff." And so all of your ability to be like, "Oh, what fields should I have access to? What permission sets should I give it?" All of that, it's all stuff that's going to make your agent more successful. Mike: That makes sense. I mean, it's also like cleaning up before you get new furniture. Josh Birk: Right. Exactly. Or maybe throwing out that old sofa. Mike: Yeah. I live in a college town, that happens every May. Every May, at the end of May, you drive around, and people can go sofa shopping because all of the college kids are moving out, and all of the target futons, and IKEA futons are at the corner. And you realize how hard college kids are on furniture. Josh Birk: They're very hard on furniture, and they usually get it secondhand in the first place. Mike: Right. They got it last year off a street corner. Josh Birk: Exactly. And became anew. Mike: Awesome. Josh, thanks for coming on and helping us understand our apartments, and understand that we should be blueprinting them. But more importantly, I mean, the cleanup of data doesn't just stop there, it's also the cleanup of metadata too. Josh Birk: Yeah. Like I said, remember your training, and I shall be successful. Thanks for having me, man. This was a lot of fun. Mike: So that's a wrap with this chat of Josh Birk. Huge thanks for coming and unpacking the foundation of Agent Success. I can't tell you how important metadata is, so be sure to check out his blog post that he's got coming up. And remember, your data structure and metadata aren't just back end details, they're the keys to unlocking smarter, more helpful agents. Whether you're prepping for Dream Force, or just organizing your org, start with a blueprint. Now, if you found this episode helpful, do me a favor, share it with some of your friends. And be sure to hit that subscribe button for new episodes downloaded. And with that, until next time, we'll see you in the cloud.

Today on the Salesforce Admins Podcast, we talk to Amit Malik, the Content Portfolio Lead for AI within Product Education at Salesforce. Join us as we chat about how admins should approach learning Agentforce and bringing AI to their organizations. You should subscribe for the full episode, but here are a few takeaways from our conversation with Amit Malik. Learning AI starts with filtering out the noise We often get asked where admins should get started with learning Agentforce, so I brought Amit on the pod to get the inside scoop. In his role as a Content Portfolio Lead for AI in Product Education at Salesforce, he’s in charge of planning the courses that are offered globally about Agentforce and Data Cloud. For Amit, the challenge with teaching AI is what he calls the “knowledge explosion.” There are so many different things that Agentforce can do, and that list is growing daily, so it’s hard to know where to get started. What’s needed is “knowledge distillation.” So the key to learning Agentforce is to focus on the core concepts of how AI works before getting into the specifics. A framework for building with Agentforce Amit goes through five questions you should ask when you’re thinking of building a solution with Agentforce: Is an AI agent the best way to solve this problem? Would it be easier to build a flow? Just because you can solve something with Agentforce doesn’t mean you should. What agent type do you need? Salesforce has several pre-built agent templates for specific use cases, like Service Agent, Employee Agent, or Guided Shopping Agents. Consider those options before trying to build something more complicated. What topics do you want to assign to this agent? Define the set of business problems you want your agent to solve. There are standard pre-built topics like FAQ or escalation, but you can make a custom topic if needed. How will you provide data to your agent? AI is only as good as the data you provide it, so you need to make sure you have everything you need in Data Cloud and set up access with the Agentforce Data Library. What actions do you want the agent to perform? “This is where the magic happens,” Amit says. There are four types of actions: Flow, Apex, API, and Prompt Template. Learning Agentforce is about understanding the layers you’re working with. As Amit explains, an agent is really an aggregation of the topics you decide it can solve. Those topics can be broken down into the specific actions your agent can perform, which it does based on the data you give it access to via Data Cloud. The art of learning is to become curious With twelve years of experience as a Salesforce instructor, Amit’s biggest piece of advice for admins trying to learn Agentforce is to cultivate curiosity. Where many people go wrong is that they approach AI as a solution in search of a problem. That can be like trying to jam a square peg in a round hole. Once you start getting curious about the business problems you’re trying to solve, you’ll find use cases all over the place for AI. But that comes from understanding, specifically, how an AI agent can improve the experience for your users. This makes learning Agentforce simple because you know what you’re trying to do with it. There’s a lot more great stuff about learning, teaching, and working with Agentforce in my conversation with Amit, so be sure to listen to the full episode. And don’t forget to subscribe to the Salesforce Admins Podcast to catch us every Thursday. Podcast swag Salesforce Admins on the Trailhead Store Learn more Salesforce Admins Podcast Episode: Architect Courses for Admins with Amit Malik Trailhead: Discover Agentforce Trailhead: Review Agentforce and Data Library   Admin Trailblazers Group Admin Trailblazers Community Group Social Amit on LinkedIn Salesforce Admins on LinkedIn Salesforce Admins on X Mike on Bluesky social Mike on Threads Mike on Tiktok Mike on X Love our podcasts? Subscribe today or review us on iTunes!   Full show transcript   Mike Gerholdt: Welcome to the Salesforce Admins Podcast. This week, we're joined by Amit Malik, a cloud content portfolio lead at Salesforce. Guess what we're digging into? That's right. Agentforce and Data Cloud, but we're going to talk a little bit different. It's about how admins can confidently navigate AI in their orgs. Amit brings over a decade of instructional experience and delivers a fresh, clear-eyed framework of thinking about AI agents. Trust me, you're going to want to hear this framework. So whether you're new to Agentforce or looking to level up your implementation game, I promise you, Amit breaks down the essentials with clarity and care for us. Plus, we also talk about why doing not just watching is a key to learning because as admins, we do some instruction as well. So it's good to learn from an instructor. Before we start the show, just a reminder to press follow on that podcast platform that you're listening to us. That way you get new shows right on your mobile device so you can listen to them when you're out mowing your yard, which is what I will be doing after I record this episode because it is summer and the grass is growing and it is gorgeous outside. So enjoy this episode with Amit. Go walk your dog or go enjoy some sunshine if it's pretty out where you are at. So with that, let's get Amit on the podcast. All right, Amit, so welcome to the podcast.   Amit Malik: Thank you.   Mike Gerholdt: I know a few years back you were on to talk about the architect mindset, so I'll be sure to link to that episode, but now we're talking everything Agentforce and Data Cloud and Metadata and Customer 360. But for people that haven't been around listening to the podcast for three years, and there's a few of them, could you reacquaint the audience with what you do at Salesforce and your journey to Salesforce?   Amit Malik: Sure. I joined Salesforce in 2013 and I have been lucky to teach audiences across the globe on Salesforce technologies every year, and I've trained on different topics about Salesforce, like whether it's about teaching administrators or developers or consultants, architects. So I've been fortunate to talk about Salesforce aspects to different audiences. In my current role, I'm working on as a cloud content portfolio lead where I'm specializing in Agentforce and Data Cloud. So my role is to plan what kind of courses should we offer to our customers globally so that we can enable our ecosystem on Agentforce and Data Cloud.   Mike Gerholdt: Yeah, I mean, there's a lot to learn. I started using the platform back in 2006 and even trying to keep up is a lot, and so I can imagine learning constantly, there's so much.   Amit Malik: It's easy. I would say nobody cares what we know in the past. What matters is what we are working today and for next 12 months.   Mike Gerholdt: Well, there you go.   Amit Malik: So I've always retrained my mind that what I have done in the past does not matter today. What matter is what I will do in next 12 months.   Mike Gerholdt: So let's talk about what we're going to do in the next 12 months. I would love to know, this is going to be my first hard question, they're all hard, I think, not to scare you. I don't think it's hard. But I am curious because you're on a different side of the fence than I'm at. What is different about teaching AI than teaching other technologies?   Amit Malik: I would say the interesting part here is there is knowledge explosion in the current times. And when a learner is learning, he has multiple channels to learn from and there is no direction what is the right way to spend your time. So what we need is knowledge distillation. We need to tell our learners, if you only have 10 hours, what should you learn in your first 10 hours. Or if you're not time-bound, what are the first 10 words should you know to do your project better in next three months. That my suggestion, that we should stop the noise and try to develop our attention span in these current time of knowledge explosion.   Mike Gerholdt: I really like that because there is a sense of needing clarity when there is this much noise out in the world. Do you find when individuals come to your classes that they have a preconceived idea of AI that you might have to work past?   Amit Malik: Yes, because the interesting part is that, for example, let's take a word AI agent. If someone is learning about AI agent from Google's perspective or OpenAI perspective or how Claude thinks about it, how we think about it at Salesforce, we may have different explanations, but at the core, AI agent is an AI system. So once we start getting a clarity of thought that where are we heading, we are trying to suggest how do you build a digital labor who will work along with the human at present. Once you know the goal, I think then it does not matter what is your understanding. You just need to understand why are we learning AI agents. Just because people are saying about it? Or do you really believe in the spirit of, "Oh yeah, it makes sense. If I can get my job done autonomously by a AI system, I can be more productive than I am." I think that's where, the why has to start first before you start doing what is what.   Mike Gerholdt: Yeah. Yeah, I often think a lot of times solutions come before people see problems and so they make up problems for the solution when in fact they didn't need to make up the problem. They already had it.   Amit Malik: Very well said, very well said. I would say in a different way that for every problem, Agentforce is not the answer. So we need to find out a use case and say, "Oh, this is fit for predictive AI, this is fit for generative AI, this is fit for autonomous agents." So when you start as an architect, you start questioning to yourself, "How should you think to arrive at the solution that, oh, this is a use case of Agentforce now."   Mike Gerholdt: Yeah, absolutely. So as somebody that has instructed admins, architects, probably even developers, I mean, you run the gamut, because I know I've signed up for developer courses. If you had to describe Data Cloud to an admin and give its relevancy, what would you say?   Amit Malik: Very good. So think of Data Cloud as a portion of all your data from multiple data sources. It's a one repository where we want to bring data from multiple data sources. It could be Snowflake, it could be Databricks, it could be AWS, it could be any data source in your enterprise business. And once we have all this data in Data Cloud, that can act as a knowledge for your agent. As the analogy, think like this. When you join as a employee to a new organization, you need to learn about the system. In the same way we want to give the knowledge about systems to our agent so that they can understand the business and then talk to the customer on our behalf about our business.   Mike Gerholdt: And I even think beyond just the customer, I mean, even the employee. If you only set up an agent to look internally within your CRM data, that's almost like giving them a book, whereas if you use Data Cloud, it's like giving them a library. Is that a kind of a competent analogy? Would you use that?   Amit Malik: Yes. So see, Data Cloud is in simple way exposing all the data sources and how we attach it to the agent technically with the help of Agentforce data library. So we do not want our learners to get worried about Data Cloud so much because in the future or in the present world, we are trying to encapsulate the plumbing of Data Cloud. So if you're a [inaudible 00:08:50], we just ask you to upload the PDF file to Agentforce data library, and then we take care of all the plumbing behind the scene. So as the admin, you just need to know how to upload your data and we take care of everything. And that's the beauty of Salesforce platform that we try to make things simple because our engineers have done a complex job for you.   Mike Gerholdt: The irony is it's usually the businesses that make things complex with processes or approvals or reviews.   Amit Malik: That's fine. See, we cannot change the business. I think as a technical person, what I've learned over the years is never challenge the business, rather adapt to business.   Mike Gerholdt: That's very good.   Amit Malik: Because that's a very important learning. If you keep on challenging, because as a technical mindset, we always think, "No, business is not right," but hey, business exist and that where technology exists. So we need to adapt to the business needs and that's where as you grow as an architect, your fight is not to learn technology because you will have lot many people under you who knows much better than you. But your job is to understand what business value business need to create and how do you bridge that gap with the technology solution. That's where the real fun is. You tie up the business outcome.   Mike Gerholdt: And I feel like, because I was just going to ask you, back to our first question, how do you kind of filter out the noise? I feel like your previous answer is kind of that, right? Really focusing on what the business outcome, the business need is, right?   Amit Malik: Yeah. So I would like to give you a very small framework which I have been using for myself as I've been learning Agentforce from the front, and I'm lucky to be part of all the teams here whom I interact with internally.   Mike Gerholdt: I love frameworks.   Amit Malik: The first question I would like to give to my learner is ask yourself, do you need AI agent? If the answer comes yes, that yes, you need AI agent, then move forward in Agentforce thinking, because it could be that you can handle something without AI agent and you don't need it. Once your first question is answered to yes, then you move to the second question, what agent type do you need? For that, you need to know what agent types are offered by Salesforce. Like we offer, say, customer agent, which is implemented through Agentforce Service Agent. We offer employee agent, which is for employees. We offer sales agents like SDR agent for initial outreach and booking meetings. We have coach agent, which is for coaching and mentoring. So once you understand our offerings, then you see, okay, given a use case where customer says, "I want my agent to answer business questions and handle my refunds, or tell about my loyalty balance," we know that I need something for service, I need something for customer, so I need a customer agent. So once you start reasoning that I need a agent and I need a customer agent and I can implement customer agent using Service Agent, then comes a third question, what topics do you want to assign to this agent? Now topic is a very interesting vocabulary which we have launched in Salesforce ecosystem, which is how do you define the job to be done? So think of topic as a aggregation of your business process. So if you want your agent to work on, say, loyalty balance check. So that's all the questions about loyalty balance will be handled by loyalty management topic. So that loyalty management topic is like a grouping of all the actions. Once you know the right topics that do you want to leverage the existing topics like standard topics, like say general FAQ or escalation, or do you want to make a custom topic, your third question is answered. Once you have answered the third question, the fourth question is, how will you provide data to your agent? And here where your Data Cloud come into play. Because agent is dumb. Agent does not know anything about your business. So to provide the value to your agent, you need to connect your agent with the data library. And now with Agentforce data library, we support web search, we support files, we support your knowledge articles. So once you understand how do you want to educate your agent, your agent is now having a data about your business. Once these four questions are answered, the fifth question is, do you just want agent to answer the questions or do you also want agent to act upon it? Customer says, "No, I want agent to act upon." And then comes the fifth thing, which is what actions do you want agent to perform? Now this action is where the magic happens, and this is where all the administrators and developers can start working on leveraging their past knowledge. So if you're good in flows, you can have a flow agent action. If you are good in Apex, you can have Apex agent actions. If you're good in API, you can have new soft calls, you can call external services. If you're good in prompt engineering, you can use Prompt template. So depending on the use cases, we give you four reference action types, which is Flow, Apex, API, and Prompt template. Once you have answered to this five questions, now you know a lot about solutioning. So this framework will help you through how to position the right solution for your customer.   Mike Gerholdt: I really like that. Thanks for sharing that. And I was thinking through as you were talking through that because we have, I mean, internally we use Salesforce on the admin team to manage our content. And Josh Birk built what we call Agent Goat, which helps us answer questions and create relevant records. We even did something fun, I don't know if you think this is cool, we think this is cool. We uploaded the release notes to Agent Goat and gave that as a resource.   Amit Malik: Wonderful.   Mike Gerholdt: Yeah. Because then we can ask it about new features.   Amit Malik: And recently we launched our Salesforce documentation action. So we can just plug in that action to our agent and now it can start answering questions based on our Salesforce documentation. We just launched.   Mike Gerholdt: I mean, there's so much to do.   Amit Malik: Yes, that's the fun thing. Once you understand that agent is nothing but aggregation of topics and topic is nothing but aggregation of actions, and actions can happen on your data. So agent, topic, action, data, that's a master framework.   Mike Gerholdt: Right.   Amit Malik: Once you start thinking on this four words, agent, topic, action, and data, everything will start falling into place.   Mike Gerholdt: So I'm curious, and I genuinely don't know the answer to this, but have you been to any of our events and happened to see any of the agents that customers have built that you find really inspiring?   Amit Malik: Yes, I was lucky to be part of TDX Bengaluru Hackathon.   Mike Gerholdt: Ah, yes. Tell me more.   Amit Malik: It was so wonderful to see our customers showcasing the real use cases. Let's say a job application agent, some student is applying for a job, that how we can have a agent which can read the resume and do the shortlisting and send the further steps for the candidate. Or it could be other scenarios which customers were showing in insurance that how we can process the claims with the claim agent. Or as we know in the Conquer, for example, expenses, how we can have expenses being reviewed by agent and auto approve, saving the time of senior management to approve the expenses. So see, once you start thinking of use cases, then the value is derived. The value is not in technology. The value is in applying the technology to benefit the customer and improving the customer experience.   Mike Gerholdt: And would you also extend the customer for admins and architects and developers to users of Salesforce, not just external customers of your organization?   Amit Malik: Yes. So let me define this. So here, when I say customer, so customer is, say, Marriott who is buying our Salesforce license, or any customer who buys our Salesforce license. And then in that customer, we have employees who are using our product, they're our end users, and if this customer is further implementing our solution for their customers, then they become end user.   Mike Gerholdt: Yep, that makes sense. Amit, I'm curious, in the instructional side of Salesforce and having done instruction for a long time, if you were to look into a crystal ball a year out from now, how do you see AI influencing instruction at Salesforce?   Amit Malik: See, it's very funny. I would say this. Everyone believes that in the future we all will learn through ChatGPT, Claude, and we'll keep on learning through AI tools. But to me, there's a need for clarity of thought. Because if you don't know what to ask, you will never get the answer. So the magic of learning is your curiosity. So as everybody's moving, we are moving towards AI teaching us step by step, but if you do not know how to leverage AI, then AI will be useless to you. So the art of learning is to become curious, and once you know how much curious you are, then AI or human does not matter. Then you can learn from AI also because you know how to ask right questions. So I think, in my mind, education is not about the instructor. Education is more about the learner. If learner is good, he can extract more from a teacher. If learner is not curious, then teacher or AI does not make a difference.   Mike Gerholdt: That's a really powerful point.   Amit Malik: This is my experience of teaching for last 20 plus years that I've seen some interesting audience and it changes the whole class when you interact with the different kind of groups. Just a good example, the way you are asking me questions, that is a prompt to my mind and I'm giving you live answers. So you are getting best out of me because of you, not because of me. So it's just a very interesting phenomenon to understand that it's all about how much you extract from my mind. My mind is prepared.   Mike Gerholdt: Yeah. I never really thought of it that way of when you go in, because as a customer before joining Salesforce, I've taken quite a few of our, I believe five day or four day classes, instructor led. One of those was with Wendy Braid.   Amit Malik: This is amusing.   Mike Gerholdt: They were all great, but they all, to your point, all of the classes were different. And they're different, well, one of them just because the instructor is different, but because the people are different and the goal of what the people are trying to get out of the instructor changes, and it also changes based on the level of curiosity, I think you said that, of the learner, right? So if they're genuinely curious, they're going to get more out of the learner than if they're just kind of sitting there expecting to be brought up to speed, right?   Amit Malik: Exactly. I would like to add one more point to this discussion here.   Mike Gerholdt: Oh, please.   Amit Malik: So the role of the instructor is not information sharing. Information is available in abundance through internet channels these days. So the goal of the learner is to help learner connect to their business goals. So when a learner is listening to the instructor, instructor need to understand what is the goal learner trying to achieve in their business goals. And you only get to know that when you're in the classroom. Before that you don't know. So how can you plan your speech? And that's where I believe in dynamic speeches where the real agenda gets opened up when you start engaging with the real audience. Topic remains same.   Mike Gerholdt: So I just had a question pop to me that I really thought brings this kind of all the way 360. One of the roles that admins, developers, and architects play is your role. They have to be an instructor to end users. When they roll out a new feature, whether it's a new application, Data Cloud, an Agentforce agent or whatever, part of their role is to teach their end users what the feature is, how to use it, how to be productive in their role. With your vast experience in teaching and educating individuals, what advice would you give to people that kind of have to temporarily step into that role and be productive?   Amit Malik: I would say listen more, speak less, think more and process your thoughts. Articulate your answer well before you speak to the listener, because sometimes we try to propose what we know and what we want, but that is not the winning edge. The winning edge is trying to understand what is the business problem customer is trying to solve, and then within your mind, you know all the products what we offer, and then find out the best solution and say, "This is what will suit you. Use Agentforce Service Agent, and that will solve you in this use case. Let me show you how."   Mike Gerholdt: Yeah, well, that's powerful advice. You would think it would be the opposite, right? Talk more. But sometimes there's a reason we have two ears and one mouth.   Amit Malik: I agree. And I have learned, again, I'm fortunate thanks to God that I've learned this by, I've been teaching for more than 20 years now, I've learned this skill in real world that the best teachers speak less and personalize to diversified audience where every audience should feel that he is speaking for me.   Mike Gerholdt: Yeah, I like that. Thanks so much for coming on the podcast, Amit. I appreciate these check-ins and really giving us this perspective on what our team is doing to get people up to speed and really also just have admins and developers and architects think about how they can be better educators themselves when they have to demonstrate the platform to their end users. I appreciate this.   Amit Malik: Thank you. I will just say at the end, as a good practice, make sure you are spending at least 30 minutes doing something on the Salesforce org. There's nothing which is better than doing something. People are watching YouTube, people are reading blogs, listening to podcast. But 30 minute of doing on Salesforce org is the most powerful experience.   Mike Gerholdt: Absolutely. I mean, if you follow pro sports or any entertainment, all of those players practice for a reason. And it's not just to be out on the field, but it's actually the motion and going through it and keeping themselves sharp and the same holds true for us. Thanks so much for coming on the podcast, Amit.   Amit Malik: Yes, thank you. Excellent. Thank you for the opportunity.   Mike Gerholdt: So that was a fun conversation with Amit. I always appreciate instructors coming on the podcast because not only do they give us insight into instruction, but they also give us insight into how we can be better instructors. So it was great to have him help unpack the agent, topic, action, data. I know I've done a lot of Agentforce NOW workshops and I've walked through that framework before, but hearing him say it kind of helped me even make sense more, like the knowledge boxes in my head don't jiggle around. So it's making sense to me now. I love that he left us with some tools to teach, learn, and lead more effectively. And don't forget his advice, which is 30 minutes in your org beats three hours of theory. That's good thinking. If this episode helped sharpen your focus, could you do me a favor, pass along with some fellow Salesforce admins, friends, user groups. I would love if you could do that. And until next time, we'll see you in the cloud.  

Why Secure AI Starts With You: What Admins Must Know About Agentforce   Today on the Salesforce Admins Podcast, we talk to Sri Srinivasan, Senior Director of Information Security at Salesforce. Join us as we chat about what admins need to know about Agentforce and how to build secure AI experiences. You should subscribe for the full episode, but here are a few takeaways from our conversation with Sri Srinivasan. Quick heads-up before we get started: This episode may include forward-looking statements—things we're excited about, but not yet available. So please make any purchasing decisions based only on products and features that are currently available. For all the legal details, visit salesforce.com. The hidden job of AI security: admins build the brakes Sri gave a great TDX presentation about AI security and the crucial role admins play in the future of Agentforce. “Admins are key to everything that we do,” he emphasizes, “they understand everything that's happening within their environment. They know which actions, what permissions, what they do, and agents are just another avenue to expose and interact with this crux of it.” As Sri puts it, Agentforce is like a sports car in terms of what it can do with your data. But how fast would you drive a sports car with no brakes? That’s why admins are so important in the age of AI. We can build the brakes for Agentforce to make sure our agents are behaving correctly. Five questions to ask when building secure Agentforce experiences Security conversations can get very scary very quickly, but Sri boils it down to five questions admins should ask when they’re building with Agentforce: What is the agent’s role and scope? What data will the agent have access to? Which actions should be public and which should be private? Do you need to build any extra guardrails? Which channels will the agent use? The key here is practicing the principle of least privilege. And for admins, that comes down to managing permissions and profiles in Salesforce and following security best practices. Every agent runs as a user—and that user needs to be tightly scoped. Test before you trust: scaling with the Agentforce Testing Center Going back to the idea of brakes, Sri cautions that just because you built an agent fast doesn’t mean that it’s ready. Luckily, his team has been hard at work on new tools to help you make sure your agents are working as intended. The new Agentforce Testing Center helps simulate and validate agent behavior at scale—without needing a QA army. You’re also able to peek under the hood to understand why an agent made a certain choice—turning debugging into decision-making clarity. Be sure to listen to the full episode for more on what admins need to know about Agentforce. And don’t forget to subscribe to the Salesforce Admins Podcast so you never miss out. Podcast swag Salesforce Admins on the Trailhead Store Learn more Sri at TDX: 5 Easy Steps for Secure Agentforce Implementation More from TDX: Security Best Practices with Agentforce Trailhead: Trusted Agentic AI Blog: Best Practices for Building Secure Agentforce Service Agents Admin Trailblazers Group Admin Trailblazers Community Group Social Sri on LinkedIn Salesforce Admins on LinkedIn Salesforce Admins on X Mike on Bluesky social Mike on Threads Mike on X   Full show transcript Mike Gerholdt: This week on the Salesforce Admins Podcast, we're talking with Sri Srinivasan about secure, reliable AI experiences with Agentforce. Now, Sri is a leader on the security compliance customer trust team at Salesforce, where he helps customers understand and implement security best practices. Of course, before we get into this episode, be sure to follow the Salesforce Admins Podcast wherever you get your podcasts. That way you get a new episode every Thursday delivered right to your phone or your mobile device. So with that, let's get into our conversation with Sri. So Sri, welcome to the podcast.   Sri Srinivasan: Thanks for having me here, Mike. Super excited for it.   Mike Gerholdt: Well, I love the presentation that you gave at TDX, and I'm sure more people would love to hear about it too, which is why I wanted to have you come back on, because everything now is Agentforce and security is always top of mind. I've always preached security ever since I started at Salesforce. I've had, I think, Laura Pelkey on quite a few times. But that was the compass of what you talked about at TDX. But I'm jumping ahead. Let's talk about you a little bit. Tell me kind of where you got started and how you got to Salesforce.   Sri Srinivasan: Let me try to make it very sweet and sharp. So I have always been in security. I have a master's in information management specializing in security. I worked for big four accounting firms, but not doing accounting. I did security for them, data security and data privacy. Then I ended up working for a little gaming company where I really got involved in security, due diligence. Was a small company based out of Reno, but they were not really small. They did almost all gaming systems, all gaming interactions, lottery, all across the world. So that got me exposed to different systems and more specifically around fraud and how systems can be hacked to do things that they shouldn't be doing. That's where I got more interested in understanding the lay of the land of security. I spent about five, six years there. Then I got an opportunity to work for one of the biggest tax preparers in the United States. I ran their cyber fraud operations group for two years down there, and then my business teams, product teams came over to me and said, "Sri, you've been on the other side yelling at us to do a better job. Why don't you come on this side and do that?" So I spent a couple of years on the product side as well. Then during COVID, I was looking back at my life when we had lots of time at home, and I realized I've done a lot of the security functions in total audit, GRC, red teaming, blue teaming, security operations center, fraud operations. One thing that I thought I did not have was that customer-facing experience, and this great opportunity came about at Salesforce, and my role currently in Salesforce is to interact with customers. My team, security compliance customer trust, is the front-facing team for all customer-facing security inquiries around security, compliance, and trust. So that's how I got here, and I've been here for about five years or so, almost five. It feels like I just started yesterday, and it's amazing. Every time I meet a customer, I just feel excited.   Mike Gerholdt: Yeah. I mean, if I had to go back in time and pick a career in tech, I feel like security is one that you're always going to have a job because if there's a lock out there, I promise you there's probably somebody trying to break it.   Sri Srinivasan: Yep. I hear you. And the frustrating part about is that it's oftentime not people trying to break the lock. It's just people forgetting to lock their locks, and then figuring out like, "Hey, how did somebody get in?" Well, you didn't lock it at the first place.   Mike Gerholdt: Yep. Oh, man. Speaking the truth. So it feels like there's kind of two eras. Well, I mean, we talk about different waves at Salesforce, but to me there's the pre-AI era and then there's the post-AI era. And for a long time, up until I saw your presentation, I kind of didn't think about security with AI, because most of everything that we do on the platform is just so secure, but let's talk about what your presentation at TDX was. So kind of in a nutshell, bring us into that presentation and what you talked about.   Sri Srinivasan: Sure. I think what the intent was, AI is the hype word right now. So everybody's talking about LLMs, everyone's talking about how to protect those LLMs, but that's just the tip of the iceberg. There's so much more when it comes to implementing AI right. Salesforce provides you with a very secure platform, but is only as secure as you implement it. So that was kind of the crux of the presentation where we actually articulated the shared responsibility model in terms of what is expected is you as a customer, you as an admins, what are the few five or 10 things that you want to question anybody in your organization that is wanting to come up with an AI solution? And we wanted to break it down from a business case perspective, in a sense, if you look at all of our top tracks around Agentforce, we break it down into role, data, actions, guardrails, and channels. Those are the things that your business users are very familiar with. If we can build security into those aspects, by nature of it, we're building security into the product itself, rather than coming at the end and saying, "Now I'm going to do a security review and I'm going to add security on top of it." So that's what we were focusing on during the presentation. Things around being very cognizant on what is the role of the agent, what is the scope of the agent, what will it do? What will it not do? What data it will have access to, and where is that data coming from? Do we need to bring that data into the Salesforce system? Do we need the agent to have access to that? Other critical things, such as least privilege, access controls, designing your actions securely. Those are the things that we spoke about during our presentation, most of which, if you just took it out of context and put it in a paper, none of this should be new words. All of this is standard security practices, but the way it's applied, the lens through which you look at it, is a little different when it comes to Agentforce.   Mike Gerholdt: Yeah, I think it's always interesting as we delve into new tech when you think of security as really telling the agent what it should and shouldn't do. Wow. Because most people, I think, probably encapsulate security into permissions and profiles and data access, but also what it should and shouldn't do is also security, right? One of the things you mentioned in your answers was guardrails. And I'm wondering, can you give us some examples of what could go wrong if guardrails aren't set properly?   Sri Srinivasan: Yeah. So I'm trying to give a better example here because I don't want it to be something that is future-looking, but rather what's in the product today. So when you give your agent guardrails, a simple one could be your agent instructions. Under no circumstances should I ask for your email address or customer name or your order number because I have all that information. If I can validate Mike, I have all that information. I shouldn't be asking you to give me that information and assume that is right. So that's a very simple guardrail that you can throw into your system, right? Another set of guardrails could be you shall not perform these actions without having the user verified. You need to know who the user is before you can go and reset his password, or you need to have their second factor. You need to do a step-up authentication before you can trigger these actions, things of that sort. And what agent does with our Agent Builder, you can start providing these as natural language instructions. And the system would know.   Mike Gerholdt: One thing I was thinking of, so I'm going to ask a silly question, because when we talk security, I feel like I'm the person that has to ask the silly questions. So I'm going to do that. You mentioned one thing of, well, I'm going to set the guardrail if it shouldn't have access, or it shouldn't ask for the order number because the person looking at the screen has the order number. Why is that important if we're not passing an order number to an agent? Why would we withhold data to an agent?   Sri Srinivasan: So we're not actually withholding the data to an agent. The reason why we don't want to explicitly ask the user for certain information is not just to ask, but we can ask, but we shouldn't trust that information. It's the innate concept of trust but verify. I can ask you for this information, but that's not a great user experience because I already know your account number, your order number. I have all that information. But rather, what I don't have is I don't know who's on the other side of the system. So that's more important for me when I say that you shall not ask for it. The reason why I explicitly state that is because I don't need this information from you. I already have it. What I need from you is to validate who you are. Once I know you're Mike and this is the associated user ID, I have all the other information. Are you able to connect those dots?   Mike Gerholdt: Yeah, no, I totally get it now. I mean, guardrails, I guess I was envisioning guardrails as like those bumpers that you put up when you go bowling and you're not very good at it so you roll a strike every time. And it's kind of that, but it's also kind of that to make sure that the conversation and the agent is flowing in a natural manner so that you're actually being productive is the way I hear it. So that totally makes sense.   Sri Srinivasan: 100%. And what we have, it's currently in pilot, is we have instruction adherence. So this is basically our systems, our Atlas Reasoning Engine has supervisory elements that are constantly looking at those conversations and getting metrics around key aspects such as instruction adherence, coherence, how factual it is, how grounded it is. These are then used to decide how the user experience should be. For example, if there is an instruction that says you shall not ask for the password through the portal, and if the system has to ask for the password, then the instruction adherence will be low and it will be ungrounded because it's going to do something that is not grounded in its instructions. So then we can set the system to say, "Block those transactions, don't do it." So the agent would say, "Hey, sorry, I cannot help you here." Whereas on the other cases, maybe we can say, "We don't have enough information," so then we can build the system in a way that it starts asking for more information so it has all the information that it needs to help you. So these are some things that are coming out. These are our guardrails that are happening when the system's executing.   Mike Gerholdt: Yeah, no, it makes sense. I mean, you wouldn't have to teach somebody the history of math, algebra, geometry, and trigonometry if all you were going to ask them is what is two plus two? And that totally makes sense to me now. Like I'm rethinking guardrails in a completely different way now. I just learned something on my own podcast. Light bulb moment. Let's talk about that. You mentioned you're talking with a lot of customers. Tell me about, you don't have to be specific, but what were some of the aha moments that stood out when you were working with admins or customers and they finally, I don't want to say finally, but there's always that moment where you finally make a really good egg and you're like, "Oh my God, I know how to fry eggs now," using that as an example because I'm cooking eggs, but can you kind of give me that, because I feel like I get to see it a lot with some of the workshops, but it's probably a little bit different for everybody.   Sri Srinivasan: So off-lead, one of the biggest aha moments that I have experienced with admins is I do run these AI workshops at these world tours, and it is really eye-opening for them to look at the Agent Builder in the middle section. When they start looking at the reasoning, they now know why an agent does something that it did. So what are the biggest reasons why this whole agents are a little complicated and different is under the hood, agents use LLMs, right? And we all know what LLM is famous for, right? They're non-deterministic. What do I mean by being non-deterministic? By non-deterministic, I mean that the same input can give you different outputs at different times. And earlier, like I think about a year and a half ago, one of the bigger problems with LLMs were they hallucinate. It's still a problem, but we have figured out how to solve it. We provide it with more data, we ground it with more truth, so that it is then working within this construct. We have RAG, we have a lot of other things that we have actually provided to solve that problem. But the other problem of being non-deterministic is still there, right? And that is why when you start looking at the Agent Builder and you can start looking at the reasoning sections, our Atlas Reasoning Engine is basically telling you there which topic did I choose, what was the utterance that was provided. By utterance, I mean what the user typed. What topic did I choose based on the utterance. And once I chose the topic, what action I chose and I executed the action. But before I execute the action, I did a plan of executing the action. If I did execute the actions, here are the guardrails, here are the runtime guardrails that I would have triggered or I would've violated. And hence, I chose not to provide this answer, or hence I chose to go on to the next step. So when admins look at it, it instantly clicks in their mind. "Okay, this is how the agent worked." And that also allows them to understand, "Oh, if I were to tweak this one word, maybe the agent would react a different way." And then they go in and they try that and they're like, "Whoa, wow. Now I've actually cracked the code of agents." That has personally been one of the biggest aha moments for me.   Mike Gerholdt: Yeah, I mean, for me, it was always, I love when I help admins and myself build a prompt, and especially when we do grounding, and then we'll bind it to a field on a page, that's always very simple, and they press this, I call it the sparkle button, and they get a response back and it's like, "Wow." And it's consistent enough, but it's not like a chatbot, right? It's not like an email template. It's a little bit different every time. But that feeling of AI isn't scary and it's not hard to do, you can see it sort of start to melt away.   Sri Srinivasan: Right. 100%. And in those same scenarios, I've seen admins go really crazy when they do the dropdown in the prompt builder and they say, "Oh, so Sri, I can actually bring in data from an Apex class?" I'm like, "Yeah, you can." And now they're able to relate AI to the things that it's dear and near to them, actions, flows, and Apex classes. Admins, that's their bread and butter, they know that in and out. So now when they're able to look at that and they're like, "Oh, it's as simple as using this in the AI world," I feel they get very empowered and they're like, "Okay, let me go play with it more now."   Mike Gerholdt: Yep. Let's touch on permissions for a little bit because I know you covered that in your presentation... Words. What are some common pitfalls? Because I know that I've gotten questions at the Agentforce NOW Tour about setting up permissions and giving people access to Agentforce, but what are some things that are just real easy things that most people trip over?   Sri Srinivasan: So one thing to understand when it comes to permissions is every time you create a service agent, those agents are running as their own designated user. We are going to be releasing employee agents pretty soon. Again, forward-looking statement supply. Employee agents actually run as the underlying user that are executing it. So if you are in your CRM panel, the right-hand side, Einstein Copilot panel we used to call it, that, now you can start interacting with it, those are kind of like the employee agents where it runs as Mike. Whereas if I create a service agent and you're interacting it through any of the different channels through WhatsApp or through an Experience Cloud, you have to designate a running user. And oftentimes, folks will create a brand new user. And good thing is that this user comes with no permissions, which is good, but the downside of it is it will not be able to do anything. So similar to your standard profiles, licenses, permissions, object and record level, all of those needs to be assigned to this user. And sometimes what folks forget, admins forget, is you have organization-wide defaults and role hierarchy that could overwrite this. And over time, these roles, these permissions, because they're like, "Oh, this doesn't work. Maybe add this, maybe add this." And over time, that role could end up having excess permission. So it's always important to review the access to this agent user periodically to make sure it's appropriate and make sure that only the right folks have access to even edit the permissions for these agent users.   Mike Gerholdt: Yeah, I always think of, especially when you're getting started, who are you going to roll it out to? Because you don't need to just turn it on and let everybody just ask Agentforce whatever question it feels because you're not set up for that. And that's probably not the business case, and that's a lot of the prep, is sitting down and saying, "So we have this really powerful tool, we have this really fast car. Where are we going to drive it?" It's not like you necessarily have to drive the really fast car everywhere you go, and not everybody needs to have keys to it. So let's end on kind of a, this not forward-looking statement, but I want to get your opinion on this. What do you think the role of admins will be in terms of shaping the future of AI at Salesforce?   Sri Srinivasan: That's a very interesting question. Admins are key to everything that we do. And admins, literally, they understand the Salesforce ecosystem. From my vantage point, I look at admins as the know-it-all people because they understand everything that's happening within their environment. They know which actions, what permissions, what they do, and agents should be considered as just another avenue to expose and interact with this crux of it. So admins should take it on themselves to make sure that we are interacting with these agents in the right way. There are right guardrails in place. And I think one thing I want to just quickly peel back, you brought in the example of a sports car, right? So let me ask you, what makes the car go fast?   Mike Gerholdt: Could be a number of things, but usually it's the motor.   Sri Srinivasan: Yes, I would say it's the brake. If I gave you a Bugatti and I told you I took the brakes off, would you drive it at a hundred miles an hour?   Mike Gerholdt: Well, you're asking the wrong person.   Sri Srinivasan: So that's where I feel the admins come in. They have to put the brakes to make sure that agents are doing the right things, agents are responsible, agents are ethical, agents are doing the right things that they're supposed to do. Yes, Salesforce does provide a lot of these out of the box, but you also need to provide. If the underlying data is biased, then the response will also be biased. So admins play a very important role, maybe not in developing all these things, but being the trusted advisors to their implementation teams to ask them the right questions.   Mike Gerholdt: I love your example, because fast is only relative if you understand slow. And fast only matters if you can stop. Great way to frame security. See why I have smart people like you on the podcast because you got to educate us more.   Sri Srinivasan: I'm happy to be here and thanks for this opportunity. And I think one thing I also want to emphasize is, just like everything, protecting data is a partnership. Salesforce provides you with the tools, the foundations, out-of-the-box topics, the guardrails. When we spoke about guardrails, we have runtime guardrails, detective guardrails, and then corrective ones too, so that we can actually look at things and make changes to the system. So we have all those things, self-improving guardrails, all of those things, we're providing that as a platform, but it's on you as the customer and admins to understand and use them. One of the biggest things that we just released was the Agentforce Testing Center. That's something that's very different, because like I said, how do you test non-deterministic outputs? Yes, I can put 20 people in front of the agent and have them test for 10 hours a day or for a month, and we would be able to cover a lot of use cases. But think about it, you've built your own agent, right, Mike?   Mike Gerholdt: Oh, yeah. Quite a few times.   Sri Srinivasan: How long does it take for you to build your first agent?   Mike Gerholdt: So that's a trick question. It doesn't take long to turn it on, but I wouldn't consider it built in that time.   Sri Srinivasan: So if we have Salesforce enabled folks to quickly churn out agents we would not be able to reap that benefit if we're going to tell the organizations, "Well, I've churned out 50 agents in a matter of three days. Now go ahead and spend 50 days per agent to test it." It doesn't scale. So that's where Agentforce Testing Center comes in. It allows you to use AI to now start generating test cases and to evaluate your agents. So that's a great addition. And as admins, we should be aware of that, and we should be leveraging that in order to make sure that our agents are secure and does what it says it does.   Mike Gerholdt: Yeah. No, I agree. I mean, does it take long to spin it up and do some configuration? No, probably faster than most. But is it ready for prime time? You can think of all the car builders. They can build cars really fast, but then they got to test them. And just because it's built doesn't mean it's ready for people to use it. So it's a great analogy.   Sri Srinivasan: Yeah, and we're bringing in things. One of the other things that I spoke about during my presentation was your need to bring in the user context. You need to do user validations and verifications. We are coming with newer features that we're going to enable agent variables. These are secure session-scoped variables that capture and store data, and it can only be set by action outputs. So therefore, this is something that you can start using to improve your trust of your agents as well. The other things you can do is that you can have filtering rules now with your agents. So basically you say that only certain topics can be unlocked if you did some other activity. For example, you can only start handling refund when the user is verified. So those are some things that are coming out very shortly. We call them the agent variables, the action bindings, and then filtering rules. These are keywords. I'm just prompting it out so that you can go in after hearing this podcast, go in, search for it. You should be able to find more details in our help articles.   Mike Gerholdt: Yeah, and our coming up events. I'm sure there'll be more stuff about that at Dreamforce. Yeah, I mean, so people ask me why this or why that? And you have to think of when you're building an agent, you're building something at an enterprise level, and it's almost like the difference between commercial grade and household grade. When you buy a mower, there's like for the homeowner that's going to maybe mow a couple times a month, and then there's mowers that are made for people that own businesses that are going to run the mower eight hours a day, five days a week, 200 hours a month, and really put it through its paces. And that's the difference between some of the stuff that we build and some of the low lift, other stuff that's out there. So I appreciate that perspective.   Sri Srinivasan: Yep. I think if we're still at time, the only other thing that I want to let folks know is we just started pilot with our Interaction Explorer, and I think as admins, this is something that we all like to do. How many times have we gone in, looked into our history or logs and we love that part, right? So Interaction Explorer takes the hard work out. It gives you all the information, one pane of observability across all your Agentforce interactions. How many users, how many sessions, what is the quality score, how many interactions per moment, what are the top ranking topics? What are the top ranking tags? By tags, what we have done is we have taken all these interactions using AI, and we've generated tags and we have grouped them together. So what that allows you to do is that now you can trace in cluster sessions with granular log data, and then now you can click down on it and inspect configurations at each and every level so that you can optimize your agents. I can actually go down to a specific conversation you had with an agent at this time, and then look at that specific interaction, that message that you typed, look in the background and say, "How long did the agent take? Where did it spend its time on? How much time did it take on utterances? How much time did it take on doing trust-related activities? How much time did it take to execute another action?" All that information is available for you so that you can take a much more smart decision on your agent enablement and also on how your agent is being consumed.   Mike Gerholdt: No, I like it. Thanks for coming on the podcast, Sri. I mean, I feel like we went through all levels and I bet if I had a two-hour podcast, you and I could talk all day.   Sri Srinivasan: I would love to.   Mike Gerholdt: There's always something to talk about, right?   Sri Srinivasan: Yes, there's definitely a lot. Again, folks know our security blogs coming out, our security sessions that happen at these world tours and other events. If you ever run into me, feel free to stop by and talk to me. I love to talk to customers. I love to talk to fellow admins.   Mike Gerholdt: Okay, I don't know about you, but my brain's doing cartwheels in probably the best possible way. Just a huge thanks to Sri for coming on the podcast. I know he worked really hard on his TDX presentation, and security and AI is always a very difficult thing to wrap your head around. But hey, showed up and nailed the seatbelt on your self-driving car and came loaded with car analogies. So I thought it was fun. And if this episode made you go, "Ooh, that's what guardrails do," do me a favor, send it to your favorite admin friend out in the community. And to do that, just tap those three dots and boom, share away. You can put it on social, you can put it on the Trailblazer Community, you can text it to a friend. And if you're hungry for more Salesforce Admins Podcast, be sure to go to admin.salesforce.com. That's where you'll find any of the links to resources that Sri shares with us, including other podcasts and a full transcript on this entire episode. So that's always good. I appreciate that. But one last thing. If you're looking to bounce knowledge off, ask questions, communicate, interact with other Salesforce admins, you can go to the Admin Trailblazer group that is in the Trailblazer Community and pop over there. There's a lot of good stuff going on. But hey, until next time, you keep those agents in line and we'll see you in the cloud.  

Today on the Salesforce Admins Podcast, we talk to Dana Walton, Senior Manager of Credential Programs and Operations at Salesforce. Join us as we chat about how the certification experience is evolving with smarter personalization, easier access, and a learning journey built just for you. You should subscribe for the full episode, but here are a few takeaways from our conversation with Dana Walton. Certifications are moving to Trailhead Academy Dana has been working on the certification team since 2015. When she started, Salesforce had nine certifications. Today, they offer 83. I sat down with her for this episode because her team is finishing a two-year project to overhaul the certification experience. The biggest change coming is that certifications are moving to Trailhead Academy. While you can still go completely self-guided with your learning, Dana and her team are making it easier to find the help you need—whether that’s an instructor-led course or curated Trailmixes and modules. Why skills are the most important factor in choosing new certifications One thing that Dana wanted to know during testing was how her team could help people figure out which certifications they should work on next. She asked Salesforce MVPs how they choose new certifications to target: are they looking for things that fit a specific role? A particular product? The answer was none of the above. The Salesforce MVPs in Dana’s testing group look for certifications based on what new skills they can learn. Armed with that knowledge, her team added a skills breakdown for each cert to make things easier to browse. They’re also adding more personalization to your Salesforce learning journey, with AI recommendations to help you plan your roadmap. Dana emphasizes that these are recommendations, not requirements. Your certification experience can still be completely self-guided; they’ve just added a helping hand. Why certification is the final step on your learning journey If she could give one piece of advice to admins looking for the next steps in their Salesforce learning journey, it’s that you need to look at every possible pathway. “Certification is not how you learn,” she says, “it’s how you prove the skills and knowledge that you’ve already learned.” Go to Trailhead, reach out to the community, or find a mentor who can help you understand what you’re getting into and create achievable goals for yourself. And then, when you’re ready, certification will be a breeze. Be sure to listen to the full episode for more from my conversation with Dana about what’s coming next for the certification experience. And don’t forget to subscribe to the Salesforce Admins Podcast so you can catch us in your feed every Thursday. Podcast swag Salesforce Admins on the Trailhead Store Learn more New Salesforce Certification Experience FAQ Certification: Salesforce Associate Certification: Strategy Designer Salesforce Admins Podcast Episode: When Collaboration Meets Agentforce: The MH4 Hackathon Story Admin Trailblazers Group Admin Trailblazers Community Group Social Dana on LinkedIn Salesforce Admins on LinkedIn Salesforce Admins on X Mike on Bluesky social Mike on Threads Mike on X Full show transcript Mike: Welcome to the Salesforce Admins Podcast. Today we're sitting down with Dana Walton from the Salesforce certification team. Dana is here to share how the certification experience is evolving. I want you to think smarter personalization, easier access, and a learning journey built just for you. We'll talk about the exciting move to Trailhead Academy and how it's making certifications more accessible than ever. Be sure to stick around. I promise you're going to walk away and be ready to take your next step in that Salesforce ecosystem. And hey, if you enjoy this episode, go ahead and give us a follow wherever you listen to podcasts. So with that, let's get Dana on the podcast. So, Dana, welcome to the podcast. Dana Walton: Thank you. I'm excited to be here. Mike: I'm excited to talk about certification. I've been talking about it ever since 2008 when I first got certified, which is a super long time ago. But let's get started with you. Tell us a little bit about how you got to Salesforce and what you do at Salesforce. Dana Walton: Happy to. You predate me just a little. Yes, but I mean, I think in terms of longevity, I hit Koa this year, so I'm very excited. Mike: Congrats. Thank you. And for those of you that don't know, Koa is 10 years. Dana Walton: Yes. So my 10-year anniversary will be in June, and I have been a part of the certification team for almost all of my tenure. I joined in October 2015 because I started as a contractor and then got hired on through the certification team, and I have been a part of the growth of this program since then, and it's been really excited to see when we first were in a very... When I joined, we only had nine certifications. Mike: Only. Only nine. Dana Walton: Only nine. Now we have 83. It's been quite a journey. Mike: Wow. I didn't know that. I'm having a moment. 83. Holy cow. Dana Walton: Through natural growth and then through mergers and acquisitions, we are now up to 83. Mike: Yeah, I didn't really think about that. I always think of the homegrown. Somebody reminded me the other day that my Salesforce experience would be graduating from high school. I'm like, "Oh." Dana Walton: It hurts when people talk like that. That's not fair. Mike: Oh, well, okay. Maybe we'll get into a good college, grow up. But honestly, I do remember that first time I took the certification. This was '08. It was a long time ago. I had to drive 90 miles one way to a testing center. I also live in Iowa, so look, it's fair. Salesforce was like, "Look- Dana Walton: It's still extreme. Mike: I don't know. We're used to it in the Midwest. They only need one spot. I could see somebody saying, "How many people besides Mike are really going to get a certification in Iowa anyway?" But I remember taking the test. I was the only one in the room. I got done. And then there's that moment where you're like, "Okay, I think I've answered all the questions." I click Submit, and then you're expecting a result, and at the time, I think it asked you a followup question, "How was your certification experience or something?" I was like, "I don't know. Just tell me if I passed or not." I was like, "It's great." And then I clicked and it was pass, and I remember I stood up and shout. Really loud. I was like, "Woo-hoo," really loud, and I kicked the chair over and the proctor ran in. I was like, "I passed." And she's like, "Be quiet." I was like, "There's nobody in here." I was so excited. Dana Walton: If you did 2008, you must have been one of our beta testers because the program didn't officially launch until 2009. Mike: Yeah, I was one of the first 500 for admin and for advanced admin Dana Walton: My goodness. Congratulations. Mike: Yeah. So very exciting. So this is what we're talking about, the new certification experience. Dana Walton: Yes. This is- Mike: Go ahead. Go ahead. Dana Walton: Sorry. I'm really excited about this product because I've been working on it for two years. Mike: Wow Dana Walton: And so we're at the finish line. We're so close. We're about to start user acceptance testing, so I'm really anxious to see how people are responding to what we've built. And the whole experience that we are building is really with the end user in mind. We rethought the entire journey of what a user would experience when it comes to how they would get started with Salesforce certifications. Where would they go to find information? And we realized that it's a really disparate journey that we offer to our users. We assume a lot on their behalf, and we wanted to roll that back and really come at it from both an expert and a beginner mindset. So first things first. It's moving to Trailhead Academy, which is the learning platform of Salesforce, and we're really excited because it brings a combined experience of both instructor-led training and certification onto the same platform. So when you go to Trailhead Academy, you are first introduced to both instructor-led courses, instructor-led training, and then you can also look up certifications, and you can start your journey wherever you choose. If you feel confident in your certification goal, you don't have to take any classes. You can go ahead and get started and register right away, but if you want to brush up on some skills or if you're net new, we also provide the learning resources on the exam pages that will help you out. And it's not just instructor-led training. We still link out to Trailhead modules and Trailmixes, and those are curated by the certification team and aligned directly to the exam objectives. So we don't forget the learner in all of this experience, but we bring it together in a new platform that then uses single sign-on through TBID so we can continue building that journey out for you as you grow in experience through Salesforce. Mike: Nice. I like that. I mean, the goal is the certification, but the journey is learning and understanding everything so that you not only have the certification, but you have the knowledge behind that certification. Dana Walton: And we're building in personalization too, so it's not just you have this certification, but it helps you figure out what's the next one on your roadmap. So if you already have started building some skills, we're going to use algorithms and AI to suggest new skills or new certifications for you to focus on next, because a lot of questions that we get from our users is, "I've done this. Where do I go to after that?" Mike: Yeah. I mean, I think that's always the discussion around certification. It's always the question I get is, "Where do I start?" Even with Trailhead modules, where do I start? Dana Walton: It's a lot. There's so much that's offered these days that figuring out that entry point can be really challenging. We have a new format for our catalog where we also highlight exams without prerequisites, so that way we're not sending you to focus on something that you're not ready to take. If you haven't earned that prerequisite, you can start at the entry level and then work your way up or stay wherever you're most comfortable. Mike: So in your initial introduction, we talked about 83 different certs through acquisitions and just growth, which still amazes me. I thought it was 18 or 20. I was like, "Oh, 18 or 20." No, 83. I'd love to know, when you looked at that whole family of how everything had grown organically and through acquisitions, what were some of the challenges or opportunities that you were looking to solve in this new experience? Dana Walton: I think the biggest opportunity or challenge that we were focused on is the fact that our audiences are so different. When you just look at it purely from an M&A perspective, you have Slack. You have Tableau. You have MuleSoft. You have accredited professional. All of those users come from a different experience for how they get certified, for how they maintain their credential, and then we have to bring them onto the Salesforce way of doing things, which is jarring in a lot of ways, because as you know with Salesforce, we like to do our own thing. We don't necessarily like to follow prescribed paths, and so trying to bridge the Salesforce method versus the history that those programs have can be really challenging for the audience. And so we're trying to be mindful in how we roll that out with the, again, challenge of the fact that we're still going to do it this way. We're going to try and bring you on slowly. Mike: You also mentioned that learning journey, the personalization, and I think that's one thing where you always have to think of... And I was a learner. I still am a learner. You're never not a learner. Where you're trying to go and the questions you're trying to ask. How does that personalization work for an individual who may only know the next year or two? Dana Walton: Well, with all the personalization, it is recommendations, not requirements. Mike: Got you. Dana Walton: And so we will look at the skills or the product areas that overlap between credentials and really say, "Okay, based on where you are, here's where we think you may be going." But it is not a requirement. It is not a checkbox that you have to complete. If you decide that, "I want to pivot and do something else," you are absolutely more than welcome to do that, and we will still support you through that journey as well. Mike: Got you. Yeah, it's always trying to understand where you want to go, because the platform is so big and so vast at some points that I don't know if there's any one person even on your team that knows everything. Dana Walton: I would not assume that at all, and I think that that's where the community is still going to be a great resource for users because we can only tell you from our experience of what we think aligns, but somebody else may out there may say, "Oh, actually, because of this really interesting job path that I took, I had these different certifications that you wouldn't think would overlap, but they serve me really well." Mike: Absolutely. I believe you said at the very beginning you're in user feedback beta testing. Dana Walton: We are about to start. Mike: About to start. Dana Walton: We haven't officially started yet, but that is our next goal. Mike: Okay. Okay. I was going to ask you what kind of feedback you were getting. Dana Walton: Well, actually, when we first introduced this project, we brought it to our MVP audience first, because before we built anything, we wanted to make sure that what we were building made sense for them. Mike: Okay. Dana Walton: Now, they are the experts, so we had to caveat that a little bit because they know Salesforce so incredibly well, and we still had to think about the people who don't know Salesforce that well. For instance, something that I thought was really interesting from MVPs, when we were trying to segment our certifications and show different filters in the catalog, one of the things we asked them was, "What do you focus on when you're filtering? Is it skill? Is it level? Is it role?" And I thought it would be role or product. And they completely flipped the script on that and they said, no, they want to build skills. So whenever they're looking at a certification, they want to know what skills that aligns to and then they'll fit it where it needs to fit. But that's where they are in their journey. And so we added a skills option. We put it on each certification page that this aligns to these skills to help bridge that gap of knowledge area that we hadn't highlighted previously. We still have the ability to segment by role or segment by product, but we added this new feature as well because it was missing from the previous experience. Mike: Yeah. Funny story: when I did get back after taking the certification, I ran into my boss's office, and at the time, as the admin, I reported to the president of the division. That's a long story, but I ran in and I was like, "Is So-and-so in?" And his assistant was like, "Yep, just go ahead. Go inside." And I ran inside. I was like, "I passed. I'm certified." He was like, "Great, what does that mean?" And if it had been a sitcom at that point, the sound of all of the air rushing out of a balloon would have been made. Because it's like, ah. But the reason I get at that is a lot of career building, a lot of what you do and skill shows through the certification. So I'd love to know how this new experience helps admins, developers, members of the ecosystem track their career and grow their career. Dana Walton: I think that is definitely something that we will focus on for the next iteration. We have not really focused on the career side of the house yet. Well, I will say that everything that we're doing will still integrate and update your Trailblazer profile, so we're not taking that away from you. When you have your certifications on your Trailblazer profile, that is still the source of truth or your certification history. But when it comes to how this experience directly impacts how that reflects on your career, we're not quite there yet. I think it is a great question and a great lens to look at, but in phase one, we're just building the brand new platform and make sure that it functions. Phase two is really about, okay, now that this is live, what else can it do? Mike: Yeah, and I love pivoting off of the skills thing, because then you can also look at adjacent skills and maybe a certification will stand out to you that you weren't previously aware. I think I did a podcast a couple episodes ago and there's a strategic design certification. Dana Walton: Strategy designer, yes. Mike: Strategy designer, thank you, that Melissa Hill Dees brought up, and I had no idea that certification existed. Dana Walton: Yes. And for instance, somebody may look at strategy designer and say, "I don't think that really applies to me," but maybe you're business analyst, and why wouldn't that apply to you? Just because you're not actually the UX designer doesn't mean that you're not involved in the strategy of whatever project or company or direction you're working on. Mike: So let me ask, because there's 83 and I can't possibly know them all. Maybe you don't. I am hearing from you that there are certifications that also include non-technical skills. Dana Walton: Not a ton. Mike: Not a ton. Dana Walton: But there are a few. Mike: Okay. Dana Walton: Yes. So for instance... Well, I will say our certification names are going to be changing as a part of this experience. Mike: Well, of course it's Salesforce. Dana Walton: Of course. Mike: We change the name all the time. Dana Walton: So I'm going to reference the old name right now, but if people are listening once the experience is live, those names may have changed. Quick caveat. But for instance, sales representative. That is not a technical focus. That is an end-user focus. It is about what key skills, core concepts, do you have as an end user in Salesforce that you need to be able to apply to your job? Mike: I like it. Dana Walton: Then, similar... Well actually, I would say business analyst does have some technical overlap. Even though we did take away the admin prerequisite, you still need to know some admin skills as a business analyst. Strategy designer would be another good example. It's more knowledge-based, and not knowledge in skills, but knowledge and capabilities that you need to be able to be successful in that certification. I feel like I'm just now going through all of the different things that we offer, and I'm trying to think. You put me on the spot. Mike: Right. Dana Walton: The majority of our exams are technical-focused. The vast majority are. Mike: But I think it's also good to know that there's non-technical ones out there to really grow that skill. Dana Walton: I mean, this is definitely something that most of your admins, I don't know if it applies to them, but for folks who are not admins but are aspiring to be, getting started with our associate certification, so we have Salesforce certified associate, that really is not technical at all. It's just understanding, how does Salesforce work? How does Trailhead work? And then if that's something you want to keep growing in, admin may be the next logical path for you. Mike: Yeah. No, I've seen that one. I think it's great. It also helps explain things when you have a friend that's like, "I hear you do Salesforce. What's that?" and they look at you. Because Salesforce is a platform, if you introduce it to most admins, they fundamentally see it as a B2B CRM platform, and, "That's just alphabet soup you threw at me." "No, here's how this works." And then once they conceptualize it, "Okay, great. A lot of people buy this?" You're like, "Yes, a lot of people do buy this." So it's an actual conversation I've had with a friend, I suppose. Dana Walton: You're right, because it's overwhelming, so it's nice to be able to break it down into digestible portions. Mike: Well, yeah. I mean, most people already assume they're going to go in overwhelmed when you start talking about technology in a platform. "Oh, I'm probably not going to understand this anyway," because for a myriad of reasons that they've told themselves, they won't understand it. You mentioned the MVPs flipping the script on you and bringing up skills. Was there any part in the design or testing process that maybe, other than that, caught you off guard? Dana Walton: Oh my goodness. I had a feeling you were going to ask something like that. Mike: No, no. Oh, you gave me the easy one. I want to find out the other ones. Dana Walton: We have learned so much. Honestly, it's the boring stuff, because as a part of this, we are bringing on a new exam delivery vendor, and I think the biggest learnings are transitioning away from a platform that we've been on for 15 years, and that's been really hard. But I think as well, we are still trying to understand, what are the pain points that our users encounter when it comes to any part of the certification experience? Where do we need to provide more clarification, more information? And so we're still learning. We are still figuring out, okay, so like I said before, once this is live, where do we go next? Because it is not just going to be static. It will keep growing and keep improving and iterating. I think that there are definitely more opportunities to integrate with more of Salesforce, especially Trailhead, to build out the full comprehensive journey for users. That is a gap that we still haven't quite evolved for. We point to Trailhead, but we don't have an integration with Trailhead yet, and so with the experience for our users and where we're trying to meet them at their needs, I think that would be our next logical path forward, but we will need to get so much feedback from audiences as well. So hold us accountable, reach out, open cases, reach out to us in the community, because we want to hear from our users. We need that feedback, because we don't want to assume we know what's best for them. Mike: Yeah. I hear you. I had a question pop in my head and I'm just going to run with it because I think it's fun. You've also been in the certification space for a long time, so I recently did a podcast with some hackathon participants, and I asked her a question, what's one piece of advice you could give admins as they're building agents? And we were talking about AI. And she said, "Well, look ahead, the next 5, 10, or 15 years. Will this still be here? And frame your decision-making around that." And that added a lot of gravity to the situation, because is it just adding a field or is it more than that? You can spin off into that. But where I'm going, so bring the car home, Mike, AI is everywhere, and we're still building tests that are very static questions, static answers. Do you, 15 years from now, see us or certification getting to a point where perhaps it's more interactive with, say, an AI agent as opposed to an exam? Dana Walton: That is the million-dollar question. I definitely think that the landscape of certification is going to change. It has to. AI is so revolutionary, it completely upends the way that people experience how they learn. There will need to be a balance that we have to put in place between memorization or recall skills, which is a lot of what certification does, versus practical application. I don't know if that practical application will involve an AI agent or if it'll involve something more like hands-on challenges in Trailhead that are in a secure environment that can be that kind of idea, where we can still see you implement skill, actually have you go through and solve a business use case in an org, for example, or some environment that is not just you answering questions. But I am also grateful that I am not on that side of the house where I have to develop that. I just have to implement what they say to do. Mike: Yeah. No, I do think it's interesting the way that AI is evolving, and we even see that with Agentforce, how you can really be conversational with it and actually give it more context and meaning than... There's been times when I've been taking a certification and I was like, "It's really A, B, and C, but that's not one of the answers, and I need to talk you through my thought process. I don't know. I'll just put D." That's usually where I end up. Dana, if somebody's listening to this and they're like, "I should probably get on the ball. I should start thinking about certifications. I should plan my career growth," if you were to give them advice on taking advantage of our certifications and all of that, what would you say? Dana Walton: I think that certification is a really great tool to prove the knowledge that you have. It is not the tool to learn, and I think that's a distinction that some people don't always grasp. And so if you are trying to learn something new, absolutely go forward and do that, but certification is not how you learn. It's how you prove the skills and knowledge that you've already learned. So if you have a goal in mind of where you want to be and you're not quite there yet, go to Trailhead, reach out to the community, find a mentor of somebody who knows that job, knows that role, or can at least help you put it in a format that is approachable and has achievable goal as you work towards that and then get certified. I think so many people just jump to certification, and that's not always the right path forward for them at that time. But once you are ready, please get certified. We love our certified audience. Come see me at Dreamforce. If you have questions, I'm there to help. Mike: Oh, that's good to know. I mean, to me, I remember 2011, standing on Dreamforce stage and talking about the importance of certification as just a common language. It's a way to, in the community, connect with other people and say, "Well, I'm a certified Salesforce admin," or, "I'm certified advanced," or now I suppose you probably have Data Cloud and Tableau, MuleSoft. Dana Walton: You do. Mike: And it's also a great way to introduce yourself in a short way that somebody goes, "Wow, MuleSoft. You're certified in that. I would like to know more. Please tell me," as opposed to trying to explain all of the tools and products that you use. Dana Walton: I think that's a really great lens to look at it from as well. It starts the conversation. It's not the end of the conversation. Mike: Right. Unfortunately, we've reached the end of our conversation, Dana. Dana Walton: Ah, look at that. Mike: So thank you for coming on the podcast and giving me a great segue at the end. Dana Walton: Happy to. This was a lot of fun. Thank you, Mike. Mike: Big thanks to Dana for helping us get through the maze of certifications and understand all of the behind-the-scenes changes that are coming in the new Salesforce certification experience. I love how the new experience keeps the learner front and center while helping you chart a clear path forward. That's something I've always wanted. If you found today's conversation helpful, be sure to share it with a friend or a team member. Maybe somebody you know is looking to get into the Salesforce ecosystem or get certified. And remember, you can always find more resources on the admin site at admin.salesforce.com and connect with us in the Trailblazer community. So until next time, we'll see you in the cloud.  

Today on the Salesforce Admins Podcast, we talk to Melissa Hansen, Co-Founder and Principal Architect at HiFi Consulting Group, RAD Women Curriculum Lead, and Salesforce MVP. Join us as we chat about her journey from fixing printers to developing an agent-powered scheduling tool in the TDX Agentforce Hackathon on Team MH4. You should subscribe for the full episode, but here are a few takeaways from our conversation with Melissa Hansen. How Melissa started her career as a Salesforce Developer Melissa started her career at a nonprofit, where she was the go-to person for troubleshooting tech issues. “You just become the person who’s best at fixing the printer, and then fixing the database, and then, before you know it, you’re a database administrator,” she says. These days, Melissa is a developer, a consultant, and a Salesforce MVP for her work with RAD Women. She’s also a member of Team MH4, and I brought her on the pod to hear what building a conference scheduling agent in 16 hours was like from the dev side of things.  Building an agent-powered scheduling tool at the TDX Agentforce Hackathon Melissa is not someone who wants to be up until midnight coding, but she was so excited about the solution they were building that it was worth the sacrifice. Like most people on the team, it was her first time making something with Agentforce, and this was a great use case to learn more about it. One of the biggest challenges for Melissa in going from building with code to grounding an agent is that the output is nondeterministic. In other words, if you run an automation, you expect to get the same results every time you give it the same data. Agents don’t work that way, they’ll give you something slightly different every time, and so you need to account for that in how you build and test your solution.  To code or not to code, that is the question We don’t always have a chance to talk to devs on the pod, so I wanted to hear what Melissa thinks about admin and developer collaboration. For her, the most important conversation to have is around automations. Flow is a powerful tool for automations, but it’s not the only game in town. Melissa’s seen her share of scary flows for things that would be fairly straightforward in Apex. For her, the biggest determining factor is who will maintain the automation after it’s up and running. As no-code tools like Flow and Agentforce continue to improve, it’s especially important for admins and devs to help each other out. There are so many more great insights from Melissa on where Agentforce is headed and how to work with developers, so be sure to listen to the full episode. And don’t forget to subscribe to the Salesforce Admins Podcast so you can catch us every Thursday. Podcast swag Salesforce Admins on the Trailhead Store Learn more Salesforce Admins Podcast Episode: When Collaboration Meets Agentforce: The MH4 Hackathon Story Salesforce Admins Podcast Episode: Transforming Conference Scheduling With Agentforce MH4’s presentation at the TDX Hackathon   Admin Trailblazers Group Admin Trailblazers Community Group Social Melissa on LinkedIn Salesforce Admins on LinkedIn Salesforce Admins on X Mike on Bluesky social Mike on Threads Mike on Tiktok Mike on X Love our podcasts? Subscribe today or review us on iTunes! Full Transcript Mike: Ever gone from changing printer ink to writing Apex code? Melissa Hansen has and she's here to tell us all about it. So, today's episode, I am chatting with Melissa Hansen, Salesforce MVP, RAD Women Curriculum lead and longtime champion of nonprofit tech. We talk about her journey from, well, fixing printers to architecting agent-powered scheduling tools and what she learned working on the team of MH to the Power of Four at the TDX Hackathon. Now, she shares her thoughts on building ai, designing for users, and what every admin should ask their developers. So, you don't want to miss this one. Now, be sure to follow this podcast on whatever platform you listen to so that you never miss an episode. And with that, let's get Melissa on the podcast. So, Melissa, welcome to the podcast. Melissa Hansen: Thank you so much for having me. I'm excited to be here. Mike: I'm excited. This whole series of talking to the MH to the Power of Four team has been such a thrill because I know we've done hackathons... by the time this episode comes out, we've done a hackathon at TDX in India. I know there's a virtual hackathon we've done. I feel like we've done hackathons everywhere, but it's such an interesting perspective because I was at TDX in San Francisco. I saw some of the teams working, but to sit down with each of you, and hear the perspective and what happened through your eyes is such a neat way of hearing the story and getting the full vibe of what's going on. So, before we get into that though, Melissa, tell us a little bit about yourself, and what you do in the Salesforce ecosystem and how you got on this Power of Four team. Melissa Hansen: Oh, sure. Yeah, I've been in the ecosystem for quite a while now. I think just over 15 years, which blows my mind. I was on a path that I think a lot of nonprofit listeners would be familiar with, which is I was in-house at nonprofits starting at the very beginning of my career and slowly moved into the technology side of things. You just become the person who's best at fixing the printer, and then fixing the database and finding the data. And before it, you're a database administrator. And I was at an organization that was adopting Salesforce in 2010. Yep, 2010. And that's how I got started. I began as an admin and I was at an organization who had some really incredible developers who were really encouraging and mentored me as I started to pick up Apex and the developer side of things. And I just loved it so much. I loved the platform and that has really been my whole career from then becoming a developer. All the way to, in 2020, after being a developer at a nonprofit for about almost 10 years, a coworker and I split off and started our own consulting group to work with other nonprofits and build out Salesforce solutions. So, that's been my overall arc. Also, I'm a Salesforce MVP Hall of Fame, and a big part of that is because of my work with RAD Women Code. Mike: Ah, a connecting Factor. I've talked to a lot of RAD Women coders. Melissa Hansen: Yes, it's such a cool organization. I was a coach in the very first round of RAD Women. Angela Mahoney, who Is an incredible connector in our ecosystem, and a lot of your listeners may know of her, invited me to coach. And I had some feedback on the curriculum and ended up taking over the curriculum as this wanted to happen. Yeah, so I've been with RAD since the very beginning. I work on the curriculum. And it's one of my very favorite things that I've done, just working with so many women and non-binary folks who are on that same developer journey I was, and who want to learn more about coding. I know this is an admin podcast, but this is our target audience for Rad, too. Mike: So, let me talk about that for a second because I feel like that's an unnecessary division that people have sown. There's admins that know how to code. Just because you code doesn't mean you have to identify as a developer. Melissa Hansen: Yeah, no, and that's a great point. And I think in Salesforce in particular, there's a ton of overlap because the declarative automation tools have become so sophisticated. If you're building some of these more complex, complicated flows, I mean you're a developer. Mike: And vice versa, too. So, how did you get connected to the other MHers? Melissa Hansen: Yes, I do all of them from for quite some time, mostly through the community. I think they're all MVPs as well. Marisa is a developer. I think she does more on the consulting side now outside of development. But Michelle and Melissa are both people who are at a lot of conferences. And we've just been really good friends and colleagues, but we've never actually worked directly together prior to the hackathon. So, that was a really cool opportunity. Mike: Yeah, it brought a lot of people together. Melissa Hansen: It really did. And I'll admit that I had initial reluctance because I am not a person who wants to be up until midnight coding. Not really my happy space. And so, I was a little, I don't know, and I've got a lot of other stuff to do. But boy, Melissa Hill Dees is very good at convincing you. And I was happy to be convinced because once we started talking about what we might build if we were to do this, we've found some really great use cases. And that's what I got really excited about actually, building something. Mike: Yeah, I mean, I talked with Melissa on previous episodes and Marissa about the conference scheduling app, which I've worked a lot in scheduling environments with Dreamforce and TDX, and it sounds like such a cool thing. And so, obviously we don't need to get into the app, but it's an agent that helps you schedule. I would love to know, because you mentioned developer. And that's the cool thing is I love that each of you had different roles in this team. And yes, some were overlapping, but because Melissa did call you out, she's like, Melissa, it sounds like I don't know who I'm talking about. Melissa Hansen: I know we didn't use last names- Mike: Melissa called Melissa the developer, like what? Melissa Hill Dees specifically was like, "Are you going to talk to Melissa Hansen?" I was like, "Yes." She goes, "Well, you got to talk to her about the developer perspective." And so, I want to start there. From your developer's perspective, what was the most technically challenging or exciting part of the agent that you built? Melissa Hansen: Well, it was the first time I really got hands-on building an agent outside of things like trailheads or quick little workshops. Which was a big draw for me because of our clients, we don't have anybody who's actively using it yet. But we want to be a little bit ahead of them and able to guide and lead when the time comes. So, I was super interested in that part. And I much more motivated when I have a real use case. I wander off if things are too theoretical or sort of like, oh, just learn it for the sake of learning it. I really need something I'm trying to get done. And I loved this use case because we have a similar one for RAD Women about matching up all of our schedules for our coaches, and our learners and our Zoom rooms. So, I could see on the other side of this some really cool applications. I think the most challenging thing conceptually is getting used to non-deterministic results. Mike: Tell me. So, before you get into that, because that's awesome. Please tell me what non-deterministic results mean. Melissa Hansen: Yes, definitely. So, typically when you're writing code or configuring a flow, a lot of times you have inputs and you have outputs, and you have the automation in the middle. And so, if I am taking in a set of inputs, let's say a few contacts, and maybe some opportunities, and I have some logic in the middle, and then I want something to come out the other end, maybe some contact roles, or who knows? And if I've configured everything, and I give it a set of data and I get a result, I expect that if I give it that same set of data again, I will get the same result every time. I could run it a hundred times, I'll always get the same result. Whereas with these agents and large language models, it's not giving you the same response every time. It's trying to come up with the right... if you could see me, right in quotes, the "right" answer for you, but that answer might change, ordering might change, the way it presents, it might change. And that is a very different paradigm than what I'm used to. Mike: Yeah, I think one of the things early on when we were talking Agentforce with customers and admins was it's not a bot. You're not plugging in if A then C, if B then D. You build upon it as opposed to having exactly follow the same path. So, now I get what you mean. Melissa Hansen: And learning how to give it instructions and natural language. When you're building out these topics, I'm starting to tell it things like, your job is to match these speakers with these time slots. And then you'd start to fill things in of you have to pay attention to the dates of the conference, and leave an hour for lunch, and make sure there's a passing period and just the series of instructions. And it almost feels weird to just be like, I'm just going to put it in a sentence. That's no syntax. Okay. Mike: Because math class, when you got to the story problems never felt like math class. Melissa Hansen: Well, maybe to me it did. Mike: I know. Math class always felt like math class to me. Don't kid me. So, I would like to know in the hackathon, I mean it's a finite time. And part of that is just you can't run a hackathon forever because you could run out of people and there's only so much that we want to see built. I'd love to know in that time, how did you prioritize what to do, what to configure, what to simplify, what to include, what to exclude. Melissa Hansen: Yeah. I mean, I think the goal that I have personally and I think that we shared as a team is we wanted to get to something at the end that was... we knew it wouldn't be perfect in a final product, but something that was really, really working. And it's fine if it's messy. It's fine if we have things we want to go and fix later, but we wanted something that was working. We obviously wanted to learn the tool set and we wanted to work together. So, those were the defining goals that told us where we wanted to go. We had discussed what our use cases were and we had a loose data model in mind, and then we divided and conquered. We had Michelle and Melissa Hill Dees were working on loading data from their respective conferences. And Marisa was participating in that so that they could be getting the data model and the data loaded together. Well, I was working a little bit with Melissa on actually building the agent. What does this thing even look like? The time crunch is real. One of our first questions was like, okay, how do we get the data once we've got it in the org accessible to this agent so that we can ask it to do all this stuff? And I think I started to build a flow. I don't build nearly as many flows as I write Apex. Mike: I mean, that's what I think every time I go to build a flow, oh, I feel like I've not built a flow in, I don't know, 16 minutes now. Melissa Hansen: It's different. So, I tried for a little bit and then I was like, given the time crunch, you go to the tools you already know and that you're fastest with. So, at a certain point I was like, I think what's going to work best is if I just build some Apex methods that pull the data query for the data, filter down to what we want, and then conserve that up to our agent. And one of the things I'm curious about as, because we're going to continue to develop this thing and hopefully get it to the point where it's a useful product for lots of conference organizers. Mike: Oh, that would be cool. Melissa Hansen: Yeah, I'm super excited. So, we have a little bit of a roadmap here, but I've got a couple of parts of it that I'm most interested in or that I have the most to do. And one is how much of the work I did was unnecessary, because you go to your comfort zone. I was like, let me write Apex that does this, gets this data, feeds it back. I was like, did I even need that? Can I configure the agent to get the data on its own? Or can I do it in a more lightweight way so that it's more flexible? Because I definitely went in with my own biases and tool set, and as I was building the agent, I was like, okay, I would think about this differently next time. Mike: Well, that leads me to the one question I was thinking about, which is... and I always do this with everything I build. There's always one part of it that I'm still thinking on, that I'm still like, if I had time, I would do X. One part of that are you still thinking about? Melissa Hansen: I'm thinking about the user interaction piece. So, our final product was a conversation that you were having within that sort of agent panel. And you say, "Okay, figure the schedule out for me." And it would output a table, but right in the chat that would show you who's scheduled for which room. And you don't have enough space really in that chat window. You could ask it to make modifications, but it's not always clear what's happening. So, it's just not the right visual paradigm. And so, the thing that I'm thinking about is getting it into a flow, into a screen flow so that the user- Mike: [inaudible 00:15:55]. Melissa Hansen: Oh, sorry. Mike: No, I was just thinking this through. Melissa Hansen: Yeah. So, that the user could launch their screen flow, maybe answer a couple of questions up front, and then we invoke the agent to do the really cool automated scheduling template for us, feed it back to the flow, and then the user has a table that they can actually work with and make modifications to make adjustments here and there. I'd really be able to see the whole thing in a way that makes sense before they sort of agree and commit, and then save that scheduling. Mike: Yeah, I think one thing for us to consider is you go and listen to this podcast when I was talking to Marissa of five years from now or 10 years from now, we're going to say, "Oh, wow. So, that's like back when Agentforce, they just had a prompt window." Back when we talk about Salesforce. Do you remember when the interface was WYSIWYG, and we could finally drag and drop fields onto a page and see them? Melissa Hansen: Oh, yeah. Mike: [inaudible 00:17:00] years ago. Melissa Hansen: It has changed so much. Yeah. I was at a presentation recently and somebody had a screenshot of not just classic but classic from a long time ago. Mike: With tabs. Melissa Hansen: Yes. And you're just like, "Oh, yeah, that used to be my every day." Mike: Yeah, yeah. Yep. And that used to be new. Melissa Hansen: Yes, it was [inaudible 00:17:23] new. Mike: Wow, that's the future. I was going to joke, you mentioned your gateway to becoming a developer was learning how to change printer ink, perhaps don't do that or do. I think it's interesting, I have a friend. I asked him, "How did you get into..." he was a consultant and he's now a CIO for a company. I said, "How'd you get into all this?" He goes, "Well, when my company got Salesforce, I was really good at Facebook." And I think even another 10 years from now, it's going to sound really interesting for a bunch of people when they ask us, "How did you get into Salesforce?" I was good at Facebook. I knew how to change the printer ink. Melissa Hansen: How are these things related? Mike: Help me out. But I do think, so I came from a non-technical background. I came from a sales background. I don't want to say a lot of admins, but a fair share of admins who really embrace the non-technical side of Salesforce that use the standard platform tools, understand it at its base level and look at things differently than perhaps a would who's been trained in code or looks at the platform first. I'd love to know, what's one thing you wish admins asked when working with developers? Melissa Hansen: We'll see if this answers your question. I think the conversation, especially since flow has become so robust, so the conversation between admins and developers is often, not exclusively, but a lot of times about what is the best place for this automation to live? Because there's way more things now that really could go either way. You could do it in a flow, you could do it in Apex. And I think there's sometimes a bias on the admin side to if you can build it in flow, build it in flow. And I have definitely seen some Flows that are a little bit scary. And not scary because the person who built them didn't know what they were doing, or did it poorly or anything like that. It's just the sheer complexity and management of a flow to do something that in some cases would be very straightforward if done in code. And so, I think especially in an organization that has code already, that has the ability to support programmatic automation, having good conversations about what is... not just can it be in flow, but should it be in flow? And that is going to continue to change as features come in and shift. But I think that's a fun conversation that's ongoing. And it goes both ways because my bias is probably to put more things in Apex than I should. And so, I need someone to say, "Actually flow's getting really good at that. We should look at it again." Mike: Well, and it's the maintainability, too. If I was an admin, and I've worked with consultants before where you sit down and it's, well, who is going... at the end of the day, who's going to maintain this? Because for some stuff we could have built a trigger. They're never going to be able to maintain a trigger. And so, if you build it in flow, it's harder, yes, but it's maintainable. Melissa Hansen: Yeah. Yeah. And that who's going to be in charge of it? Is one of the biggest questions because absolutely, if there isn't a developer on staff or on retainer than taking more time to build a flow to have a chain of declarative automation. Yeah, that's going to be the right answer. Mike: Yeah. So, I'll ask you this. As a developer, how do you see AI tools like Agentforce changing what the day-to-day is for admins and developers? Melissa Hansen: I mean, it's changing so much so quickly. Most of my actual hands-on usage is with Agentforce for developers. That tool set and related AI tools that are geared towards developers have already changed my day-to-day so much. I have conversations with them about code that I'm refactoring, or writing tests. Or a very common thing for me is I'm going into a new environment and now I can have the agent summarize what is happening in these classes and give me something that would've taken me quite a lot longer to just try and pour through, and read and synthesize on my own. It gives me a good jumpstart. And you're starting to see that more on the declarative side as well, where we're going to have agents where you can say, "Summarize this data model for me," or, "How are these objects related?" And my favorite would be, what automation is happening after I save an opportunity to have something that could look across your system, summarize both declarative and programmatic automation and say, well, there's this flow that fires, and these two triggers and a field update. You're like, wow, [inaudible 00:22:50] a bit of a much harder question to answer. And it feels like that's the direction we're going in. It's like this little assistant that you have that you can ask all kinds of questions. Mike: Yeah, I mean, because there's such a big platform behind it. So, I want to ask the next level question for one of those. Internally on our team, I did a ChatGPT demo for some of our marketing people to show them, here's how I do this, and essentially, here's how I use this to make me a better writer. And one of them very astutely asked me like, "Mike, how many hours a day do you think you save?" And I don't know. I mean I use it extensively to help me write and edit better to compare. It's easy to compare big blocks of sentences and paragraphs together. What I'm getting at for you is you mentioned having it and having AI compare code and tell you what's going on in the class. How long would that have normally taken you? What's the time saving you're getting by just dumping in some code and saying, "What's going on here?" Melissa Hansen: I mean, it's definitely on the order of hours. It very much depends of course. If I'm looking at something pretty straightforward, take me a few minutes. Or just made it a lot more pleasant potentially. But for something that is pretty gnarly and I need a lot of context to even get started, it can absolutely be, I'm like, wow, that helping me refactor and summarizing things cut two hours off of what would've been a six-hour project. Mike: Holy cow. Melissa Hansen: Possibly more. And there's some days where I don't end up needing a lot of that, and things are as usual. And then there are some days where it's like, that would've taken me twice as long. Because you still need to validate with telling you. We're not at the point, if we'll ever be, where you could just say, "Refactor this class, compiled, I'm done." Mike: Right. Well, I mean that was exactly one of the use cases I was going through with our team. She pointed out, she's like, "Yeah, but that's still not right." And I was like, "I know. That's the human in the loop part." And I told her, I go, "That's where you're the human." And we kept rewriting a sentence and it just never got right. It got really good, but it just never got right. And I was like, well, that's because it's not there yet, but it got us 80% of the way there. It's like Google Maps. When you put in an address and it's like when you get here, you just got to figure out the rest of the way how to get to the front door from the parking lot. Melissa Hansen: Yeah, address is on the wrong street, but actually the entrance is over there. But yeah, figure that out. Mike: Yep. Yep. One of the questions, so this came happenstance during the Melissa Hill Dees episode. And when I asked it, I realized I got to ask everybody this question. So, I asked everybody this, if you could build an agent to help every admin do one thing better, what would it be? Melissa Hansen: That's a good question and I really need to think about it. I'd probably go to data model. Mike: Oh, all right. Melissa Hansen: Yeah, something that is aware enough of your organization, what it does, packages you've got, the objects you've got, that when you're solutioning and you're proposing, maybe we're creating some new objects or maybe it's existing objects and we're going to create some fields. Something that can ride along and really help you make sure that the way that you're architecting that object in those fields makes sense. And I would take it a step further and have it really help with the self-documentation side of that as well. Mike: Yes. Oh, my God, Somebody finally said it. I was hoping somebody would say it. You said it. Melissa Hansen: I think those labels, those descriptions, the help text. It's always been helpful and important to have those things for different people coming in and learning the organization for helping users. But as we use more and more agents, it becomes even more important because those tools are using your documentation, that metadata to understand and give you better answers. So, I can ask it, "Hey, give me all the contacts who are this and this kind of member," and it can give me a good answer if the fields that tell you what that means and what kind of memberships are, if those things are annotated and filled in, it's going to give me a really good answer. If they're not, it might be guessing and it might pull an old field from a package we don't use, and give you a really wrong answer. Mike: I would totally use your agent and ask me, based on the data architecture I've given you, can I build this report? Melissa Hansen: Oh, that's a good one. Mike: Because we all have an Achilles heel. Reporting is my Achilles heel. And for as much as I'd love getting reports out, I'm always afraid when I'm architecting an app, that I'm building it in such a way that I'm not going to be able to get the reports that I want. I would love that. So, I would love to use your agent to be like, okay, before we commit this, I'd love to have it say, "Here's the reports you can build and give me some sample reports." Melissa Hansen: And that's such a great example as a time saver too, because typically that's always a validation step after you've built, can we get the report? Can I build the report? But if you can answer that question before you implement the data model and make all those connections, you can fix things a lot earlier and a lot quicker. Mike: Or have to make all of those really crazy custom report types. Yeah. That was it. Melissa, this is a fun conversation. I'm glad you came on and gave us some perspective on the agent that you guys built and some of the things you're working on. And I had a million more questions, but- Melissa Hansen: Time is finite. Mike: Time- Melissa Hansen: Our time is finite anyways. Mike: Time like the hackathon isn't on our side. But I think it's really cool what you guys built, and I would love to see something like that come out. I think for people that have listened to the podcast and they're like, "Oh, it's a conference schedule thing." Yeah, I don't think we'd use that. Folks, it's like meetings. If you've ever got two or three people, you got to schedule meetings for this thing, there's so many relatable uses for it. Melissa Hansen: Yeah, absolutely. I can think of a lot of places where this would come in handy. And even, I think just learning it is such the case that once you get into it and see how it was built, then you start to see the other applications. It's like, oh, but we could use this for managing our conference rooms or a whole variety of other things. Once you see the kinds of questions that can answer for you, given data that would take humans a long time to munch through and align. Mike: Yeah. And you only had a finite amount of data to work with as opposed to what an organization may have after even a few years. Melissa Hansen: Right. Yeah, there's some rich data sets out there, which is great because yeah, we're going to continue to develop this thing and hopefully have a useful tool that other members of the community can utilize. I'm thinking about taking it to the next community sprint, and seeing if we can get some more help and more use cases over there because such a cool, cool event. Mike: Yeah, everybody needs to share out those use cases. Let's end on a high note. What's one word your teammates would use to describe you? Melissa Hansen: It's a hard question. Let's go with analytical. Mike: Oh, I feel you were very analytical in coming up with that answer. Melissa Hansen: That's fun. Mike: That's great. Well, Melissa, thank you for being on the podcast. This is wonderful. I feel you've done a lot to inspire people to just sit down and be curious with an agent, and see what they can and can't do, and just be creative with it. Melissa Hansen: Well, cool. Thank you so much for having me on, Mike. This was really fun. Mike: Big thanks to Melissa for joining us and sharing her thoughtful insights, not just on building with AI, but on collaborating across roles and using tech to empower others. If you enjoyed this conversation, share it with a fellow admin or developer. And don't forget, you can always find more resources at admin.salesforce.com, including a transcript of today's show. Be sure to connect with other Salesforce admins in the Trailblazer community. And as always, until next time, we'll see you in the cloud.