On this week’s show we speak with Railsware CEO Yaroslav Lazor about moving from being a business building software for others to being a product company.
You can find out more about their BRIDGeS framework here: https://railsware.com/bridges-framework/
Yaroslav mentioned Railsware’s products:
Coupler.io: https://www.coupler.io/
Titan Apps: https://titanapps.io/
And they’ve listed some trusted resources to support Ukraine here:
Show Transcript (produced by Descript, so potentially with a few errors)…
Matt: Hello and welcome to episode 279 of WB40, the weekly podcast with me Matt Ballantine, Chris Weston and Yaroslav Lazor.
We are back again. Welcome to the, I think three, but, but to the end of the year, because already we’re hurtling to 2024, which is a terrifying concept. , Chris, how has your week been?
Chris: I’ve been traveling around this week, Matt. Rather unusually, because it’s been quite quiet. A few weeks for, for travel, but I actually went to London this week and I went to the launch of the open data.
, The latest report on open source, the start of open source and, the invitation of Amanda Brock, who has been on this podcast a couple of times. And hopefully we can get back on again soon because that’s the open source world is, is one that needs, , , a lot of support.
. It was really interesting actually, they’ve actually created a little video of what’s going on in terms of open source and the kind of things that it’s promoting in the UK and they had the Shadow Minister for Tech and Digital talking there about, she seemed to get it, which was good, although I noticed in the news tonight they’ve had a bit of a reshuffle and she’s been shuffled off somewhere else, so they’ve got a new Shadow Minister for Tech and Digital.
And, and I also spent some time with, old colleagues and new colleagues. I saw, , a demo or not actually a demo. I mean, it’s a, it’s a product, , of a, of a service that uses, one of the LLM models, I think Claude or something like that to create training courses, right?
So you tell it you want a training course, you tell it how many , modules you want, you tell it you want. Questions at the end and it goes away and it does it and it builds it into a, what do they call them? Something like Moodle, LL, LM, LL.
Matt: Learning Management System, LMS. Yeah,
Chris: LMS. There we go. Uh, all those L’s and M’s are confusing me.
So, uh, and it builds that into AutoMagically and, you know, that’s quite an interesting use case. And I attended the, uh, Worship World Company of Information Technologists. Dinner as well. Lunch. That was, which was nice. Met, some new people there and, , had, , the company of,, a lady, , who works in the cabin office and also James Boar, our, , security, , guru who stalks the podcast.
So, yes, uh, it’s been a good week, Matt. It’s been an interesting and, and, and very, , varied week. How about you?
Matt: , I’ve had far less. , entertainment, by the sounds of it. , , we did the, , keynote karaoke last week, which was great fun. Seemed to go down well. , I think in the end eight people volunteered to be able to make up a presentation to unknown slides as they went along, which is, , entertaining.
And I’m now plotting to see if we can do it in, , other formats in other places. That’d be good. And just chucking away, it’s that kind of making sure that stuff is, particularly with clients, everything that we have needs to be done and dusted before the Christmas break gets done and dusted before the Christmas break, particularly making sure that contracts needed to be in place for the start of the new calendar year are in place.
I’m interested on the open source stuff. What is, what is on the agenda for open source in open UK
Chris: Well, they’re very keen on understanding how much people in the UK contribute to open source and how much influence we have on the development of those products. So they’re looking at these kind of mapping commits, mapping, things that have actually been accepted into the, the open source software around, around the UK.
, and trying to, I think from my point of view, what, The question is, how do you then take those people who are, imaginative enough and motivated enough to support open source projects, who could then become part of success stories in the UK and stop them just being hoovered off to, , you know, the Bay Area or something, because they, you know, they get a job with one of the American companies, right?
And that’s kind of the… That’s what the Take UK or Open UK are trying to achieve with this, I think is to, is to protect and, and nurture some of that, uh, talent for the uk, which I think is
Matt: reasonable. No, absolutely. Um, and is there any sort of trend within any of that? Is there more open source activity going on?
Less is it level or,
Chris: uh, that was something that we didn’t get any numbers on at the time, but they are working on that apparently. Okay. So that should be sort of a later thing. I did ask that question actually. Um, I was interested in how. relative to the code base, basically, that’s that, you know, it, is it growing, but are we falling away relative to the code base or, you know, but yeah, those numbers are still being
Matt: crunched.
Difficult numbers to calculate as well. Cause you do run into that risk of, um, working out on metrics that are based on number of lines of code created, which of course is well, exactly
Chris: how much is being used. It’s, it’s, it’s a, it’s a bit of a. A bit of an art, I think, rather than a science, that one. Mm.
Matt: Absolutely. , talking of the Bay Area, , Jaroslav, , how are you doing?
Yaroslav: I’m doing pretty well. Thank you for asking. It’s not a Bay Area here. Los Angeles, and,
Matt: but… We’re a bit, a bit down the coast.
Yaroslav: Yeah, a bit south. , doing all right. , wanted to contribute quickly to the open source discussion you guys were having.
, , it’s interesting that UK as a government is thinking about it as it’s considered as, well, kind of a currency, right? You, you have, it kind of is, right? That there’s pounds, but there’s also like lines of codes that people contribute. And , , it’s one of the places where , engineers can, can be hyper creative because they are product managers within.
The product managers, product owners, engineers, , within what is it that they do, uh, in the open source. As it’s kind of used for this internal market, uh, engineering market, right? And then, and they are utmost creative and passionate about doing that because you’re not being paid. So, , it’s a bit tricky to monetize that or even to use that to some extent.
, but it’s a, it’s a great passionate energy, uh, for, for those folks who, who do. , those contributions, to to put them to good use. But we found that it’s very tricky to use that and turn that into productizing. From some perspective, like, I mean, you take people who are great at open source and then you put them on the products, but then they don’t have the same amount of ownership or passion or interest because it’s a very natural interest that they had before.
And now you kind of have to, you kind of have to use it. Um, in some, some other, you know, space and like my daughter is, uh, she’s an art student and she, she’s going to high school and she said, well, they making me do art by the deadline and this is not how I want to do it. I’m an artist and, uh, and this is a similar thing.
I love to persuade people to, to be productive. To be product oriented, to make products better, but they, you know, tend to feel that there’s politics involved in that. Oh, I have to talk to this person.
I have to make that person change their mind. And here I kind of didn’t have to do it because I’m, I’m an artist. To produce this open source. So it’s, uh, it’s tricky to take those artists and, proceed them to change humanity for the, for the better.
Chris: That’s, that’s exactly how I used to be when I was a, when I was a coder.
We weren’t even called coders. Then we were called programmers. And somebody would say, we need to do this by the end of the month. And I’d throw my hands up and say, you can’t ask. It’ll happen when it happens. I can’t be responsible for the. Artistic process. This is, this is a bad inspiration. I can’t be sure when it’ll hit me, you know, and, uh, in fact, I’m pretty sure there’s plenty of programmers out there just like that now.
Yaroslav: Yeah, I think people should surrender to this as well because when you’re a hobbyist, you do something for the hobby and you have this urge to do something, it’s great. But when you do something as a professional, you know, you show up every day. To turn on this, kind of inspiring process within you and actually be creative, within certain boundaries.
This is what like made human, human race evolve and survive, , throughout times. Right. And then, I think people should just try to try to do that. I will definitely persuade my daughter to, to be creative in, in a, in a calendar because it’s, uh, it’s just, , I think all the best things that we have in the world, uh, music that was created by, by artists who were doing it over and over and over again.
They definitely, uh, they were definitely doing it kind of on the schedule, repetitive, they were pushing themselves to do that.
Matt: . There’s also so many stereotypes on there about , , the tortured soul of the artist and it’s all to be incredibly hard and difficult and when the creative muse hits and again, I’ve worked with many developers over the years have been like that as well.
, but it’s, it’s a process. And one of the things that is. , it’s almost like a, revealing of how the magic trick is done is when people actually start to be able to talk about how the process of creativity works. There are a lot of people who find that very disturbing because they want it to just be some sort of magic thing.
And there’s, there’s a balance there and there’s, who knows when inspiration strikes, but actually getting the stuff done is often about being able to, as you say, being able to do it to the calendar. Learning those skills is really important. , so anyway, this week we are going to be talking about, , building software.
We’re going to be talking about transition of organizations from different modes. And so I think we should probably crack on,
, businesses go through different phases. They have to be able to adapt to the spaces that they’re in. They have to be able to spot where there are opportunities to be able to do things differently. And. Well, I think we’re going to find some interesting stories on this week’s show around exactly that.
So, Yaroslav, do you want to just give us a, a bit of a, um, a potted history of your business, a little bit of , um, the story of how it came to be and what you’ve done and how you got to where you are now?
Yaroslav: Our history is also a bit dramatic and artistic. , I think we gave people a lot of space and a lot of trust and a lot of, , stake in the game at the early days, uh, when we were not growing.
So the company is like 16 years old. And , we have been, , hiring and figuring out how to build software and learning from like really some of the best, uh, in the industry, , on, on the deep principles of building software. And we, we have a lot of frameworks where we would give people those, uh, ability to vote.
We call this thing a product fair and we would show all the products we have and then we ask people, you know, to vote what is it they want to work on and we gave them, measurements so that they don’t have to, , write comments, , and there would be like, really want, want. You really have to, don’t want, I’ll experiment with stuff and from that we’ll be, uh, data driven and, and to try to pack the best teams possible to work on the best products.
And I think we moved away from that a bit because, , when you give a lot of,, choices to people, they get confused without a lot of information, but then when you give them a lot of information, they get even more. confused. So now we’re a more driven company. So within our artistic times, we would be building products on the side.
We will be helping clients build their products. , we built some amazing software over the years , we were like pioneering some of the domains and, and doing some great, high quality, high performance software. We were building things on the site, our own products, but we were hesitant to deliver them.
You know, they still had bugs or maybe they didn’t have a website. , we, we built a CI server, a very nice CI hosted CI solution long time before there were CircleCI or the other like top tier products. Which are probably now billion dollar companies. And if we would launch it back then, I think we would have a billion dollar company.
But we didn’t because we were artistic about it and we wanted it to be great and we were thinking, we were trying not to put a managerial push into the process of engineering creation and so on and so on. That was really stupid. Then as we were helping other companies build their software and looking and watching how they grow and how it transforms,, using all the approaches and principles that we learned, , , we saw this kind of creative process.
, one of them, the cleanest out of them is Calendly, because it was started with us only by our team. , I was a product manager in there. , there was a lot of influence, , for , how the company was started. You can start it in so many ways. And we were using our product management principles that we, we believe in, , basically to, to see potential directions, what would those directions entail?
How hard is it to do them? How hard is it to start? , and Calendly became a flopping success, right? And then after that we were like, Oh wow, we definitely need to push our products further. And, , around 2019, 2019, we started being more serious about the products and we started to do basically the same thing we were doing, but we were doing them more artistically.
Right. And then to, to become disciplined about it, actually something I just talked about on an open source software, right. push ourselves to deliver them. And it was pretty hard because you, we had to hire a very different crowd of people than, than we had before. , so we, you know, typically we have product managers, engineers, , designers, and, , QA and , in order to actually deliver the products, we need support and, Sales and design and marketing, a lot of marketing actually, , and security, , to just keep, keep things, , up and running.
And, and that’s, , actually a very different crowd. , and then you don’t understand how those things should function. So we need to dive into those domains, understand how they function. And then you, only then you can start hiring people. So this process was. This process was quite a bit of suffering and, , and it’s like going to stretch artists who stretch.
Going to a gym is one thing. It’s, it’s hard. You, you, you kind of like, you push yourself, but going to a stretch artist is pure masochist activity where they basically stretch you and it hurts all the time and you know it’s going to be a huge pain. It was very similar, , to basically to gain knowledge into all those other domains and, and to become great at it and kind of to divinism.
So,
Matt: , Did you have to, , take a different approach to the things that you’d been doing before? So you have product managers and developers and QA’s as you went from building software for other people to building software for your own products. Did that mean that those core functions needed to change mindset as well as building the other services around it?
Or was it a similar sort of approach for the, for the core and then augmenting with those other skills and capabilities?
Yaroslav: Yeah, so it’s actually very similar, but in the same sense, it’s very different because when you have a Product owner or a leader of some organization and you take him and you persuade him into a certain way, right?
So we have this process called bridges or inception where we would take the domain of the product what we’re talking about It can be anything school system in US which is very different than than anyone would used to or In insurance space, , again in U. S., most of our clients are from U. S., , or Calendly, for example, , there’s a subtle, craft of booking events and closing the deals and keeping things tight, , and , we would use our bridges, , principle to build this massive board of, , who are the people that we’re helping?
What are their issues that they’re having on an everyday basis? What are the benefits that they want to, to have? , what are the risks that they want to mitigate? What is the domain knowledge of the people that want to build the software? What is it something you know that other people don’t know?
So basically, and you build this board of knowledge, everyone is looking at the same board, and everyone is reading the same cards. And… Our brain is not meant to read big pages of paper and be understanding everything from it. It would understand something from it. Our brain is more network y, right? It’s a network of connected events.
So this is what we did on this whiteboard. We were putting one card, which meant one thing. For example, that it takes six… Six back and forths to book a meeting with someone. And if you hit the, or a do knowledge that if you hit a spot, , of booking a meeting within the first 10 minutes, then you have like a 90% chance of, , securing the deal faster than the other people because you just got someone cut in front of someone.
So the, you know, there was something, uh, as a very basic, uh, important information about gallery and then for a salesperson. What do you mean? Everyone knows that. And then everyone doesn’t know that and they think there’s more important things to it. Like, you know, keeping settings of your meeting or, something like that. And, , when, when everyone, the product managers, engineers, and designers, when everyone takes time to actually talk about the things. and outline them into cards, they have those recorded clips, video clips in their head of knowledge , , that was connected to this one card.
So every now, every time now, when you look at this card, you remember this clip and you remember this knowledge.
So Bridges gives you this ability to literally be on the same page because you see this big sheet, big page of cards that, that we’re talking about, and then you’re pointing fingers to them when you make arguments. , and then we have another principle is called the heart, , which is you’re trying to figure out what is it that you need to build to, to make it’s like physical or digital appearance in the world so that everyone starts looking at it and say, Oh, this is what we’re talking about.
I had a couple of variations in my head and he had a couple of variations in his head. But now as you start to building things, not from what’s most clear to you, but from what’s most important for the product. What are the most important forms and essences that the product needs to take shape of right and then and and then people are like aha This is this is what’s happening, right?
Even after you had all those discussions So for example, if we if we go back to one of our products, which is a atlasian marketplace Add on, it’s a smart checklist So it’s a, it’s a checklist, , and it started from a super simple thing. It had a text area where you could write minus, , checklist items, plus would be a finished checklist item, and tilde would be in progress, x would be canceled checklist item.
So it had a text area, , it was storing this into Atlassian. , marketplace as a custom field, and then it would have a saved button that would render that into an H ml, which had check boxes, the appropriate check boxes. And when you check the checkbox, what we did is we resave the whole text. And if two people would , , refresh the page, one person would keep it for like few days and then he would go and check on the checkbox.
He would actually rewrite the old version of the checkbook. Which was fine. And we did it in just a few days. And then we delivered the product. And there was a product. And it, , it was like, you know, six months in the marketplace and about 700 companies installed it. And we just spent three days on it.
And that’s it. There was a software. It gives you a lot of stuff. It gives you the ability to search. It gives you the ability to do automations. It gives you, it opened up so much stuff that we had in our, uh, in our ideas, but in order to build it as a cloud software with APIs and integrations and so on, it took us like a year of four engineers, which eventually it took, right?
Chris: That’s a really good example of. Developers building software that they see a need for themselves, right? And they build something that other developers can use. But you talked about the fact that you had to have, in order to build products successfully, you needed all these other people.
These support people and marketing and all of these. Is that? In order to, to bring those other perspectives to the developers that they don’t have
Yaroslav: yes. Yes, and you do it in inception phase You try to do it in inception phase. It really depends what stage the software is on. And it’s called Inception, but it’s actually, it’s weird when you do Inception in a software that is 5 years old or 10 years old. So, we do Bridges Session.
And, I think we as human beings are like, generally perfectionists somewhere down, deep down, right? And we want to build it, like, we want to put all the information and do everything proper. And like, do it the right way. Which, which is like, whatever. And then if you put too many people in the room, you’ll kill the dynamics of the, of the team.
, so as a leader, you have to understand, do you need to bring those marketing folks, because they might be hanging in the meeting, contributing nothing or being very confused of what are we talking about, or they might, they might contribute. So you need to understand what the dynamics will be in that room.
And you’re, you’re kind of cooking a dish as a, as a person who runs this show, you’re cooking a dish and you need to understand what to add there, right? You don’t want to add the dessert, the soup and the main dish and all of that and put it into a blender and say, here, here’s a nutritious smoothie.
Please eat that. It’s not going to be that great, but so you have to be careful on a lot of those things. And specifically the most important thing, I think, is the team dynamics when, when team is not confused and they’re happy. Excited about the journey. This is like, is one of the most important things.
So this is what you start, you should be cooking from. And, you, you need to add just enough ingredients and then explain it to the team, explain them enough of vision. Not too much, because if you explain too much vision, they’ll go like, oh, you’re a lunatic. Or if you explain to little vision, you say, well, this is a copycat thing.
Like, you’re not inventing anything, right? So you need to be very careful. And once your team is healthy, you can start building around them the next layer of team members. Those could be in support and sales, , those could be other engineers and product managers and so on and that, that was one of, this is one of our most important, , ingredients.
The proper onboarding and appropriate dynamics of the team. Just feeling of people, and, and kind of like nudging them into certain direction. And for them to have their own ownership of, um, of how, how to, how to do it. You kind of say what, and they say how, and then you, you dance around understanding that.
Matt: There was a, uh, interesting discussion, , on the, show signal group, today about the, the description of building software as engineering and actually what you’ve described there is really, I think, really interesting because what you’ve described there is actually software is almost what comes out of the social interactions that you have between people with particular technical skills, but it’s, it is very easy.
You know, the software that is produced can be incredibly influenced by the relationships that exist between the people who are producing it between the customer and the people producing it and being able to understand the way in which that kind of social fabric sits together in the way in which software is built, because there are so many options that you can take, and there are so many ways that you can do things.
And there are so many, decisions to be made that could take you in a good direction or a bad direction. It all comes down to you. At the core, the people in the relationships between them.
Yaroslav: Yeah, it does. And it’s all very tricky. All those subjects are very tricky. There, there are some rule of thumbs, but basically, and there are like, there’s a negative and positive rule of thumbs.
If your team has good dynamics and they, they respect one another and respect them. very important, , element, then, then they’re going to get, they’re going to be better, but only if they only have respect and they don’t have knowledge how to build something. But if they don’t have respect and they just want to be mean to each other and crap on each other’s, , pull requests and, and, and so on.
Then, because people are big children, let’s, let’s face it, there’s, there’s no adults in this world. Um, looking at what the, you know, the heads of states are doing in, in, in the world. It’s like, well, I’m, I’m sorry, there’s zero adults. There’s no fast way of, of doing it. You got to, you got to force through it.
You have to, you have to slowly, , build the team from the proper people, proper leaders, proper dynamic, , and be very into it. And that’s why I, most of those 16 years where my learning years, , and like, I’m an engineer. I love to write code. I don’t get to do it that much nowadays because we have, you know, 70, 80 engineers in the company who, who, who are really great at it.
, but it was the, the core skill that I got. And then we learned the principle of T shape, , which is like expert generalist, right? And then only expert generalist can understand one another. If you put a smart product manager and a smart engineer in the room and you ask them to communicate, and then you like look from this like one side glass.
You’ll see how much miscommunication and, , struggle there is, might be, right? They might just work it out, but they might be so confused. And then the program manager doesn’t want to feel silly. And he would say, ah, yeah, I get it. But he didn’t, you can see it in their face. And then the engineer would say, yeah, sure.
, but he was like, well, I’m just going to do whatever. So I have this saying that I love, which is like, engineers won’t build what you want. They build what they understood. And they will basically do it based on their dynamics. Right. And then they, they think like they have all those weird rules and wrong dynamics and other companies and so on, where people said, like, you know, someone said like, don’t touch that, like, uh, don’t touch it if it works or why didn’t you touch it?
, it could be improved. And then, and then they live in this crazy world in between, , where they need to kind of figure out who you are and what is your management style? Are you like a bad manager? Do you really understand what’s happening? And they’re really trying to be productive. Like. Play it cool, don’t be a fool, and so on.
And then they start building some things like, you know, login screen or forgot password screen, and do some super great method of super authentication there, or some new way of doing stuff. And then the product manager will say, So are you, like, progressing? He’s like, Yeah, yeah, look, it’s… I already have three screens, three useless screens that no one probably needs.
Chris: But you’re talking about, , you’re talking about experience, aren’t you?
You need people with experience and, and as you say, , a view of more than one area in order to be able to make those judgments. And I’m thinking about the. Claims that we’re currently, being made about the ability of, , large, large language models and AI powered systems to write code and how that’s going to improve, , code of productivity by 40 percent or whatever it is.
And I wonder whether that ability to. Understand or to be able to judge whether something is understood and whether people because it’s all very well having an ability to increase your productivity by 40 percent if you’re 40 percent in the wrong direction. How important will that sort of human side be in future?
Do you think it become more and more important than the technology technological side because the commodity stuff is done by the computer or will we never reach that stage?
Yaroslav: So I, so basically I remember watching this movie about, the , band, super, super popular, Queens. And there was this time when Queens separated, right?
And then the, , the soloists, , just went the opposite direction and he hired the best musicians and he wasn’t able to produce great music with them who were like, supposedly be more professional than the folks that he was hanging out with., and I’ll explain where I’m going with that. And then basically he came back and they started to produce great music after that.
And, a lot of people say, well, I’ll… Programming is dead. LLMs will write all your code. They definitely can write a lot of code, a lot of hallucinated code as well. but who will make sure that it actually is the right code that they wrote? What, , some of the more popular thoughts, are that they’re not here to replace someone.
They’re here to speed everyone up. Those people are professionals who, who need stuff done. Very often you would say like, I just need to say, I have it in my head. I just need to type it in. It’s not necessarily type it in, you type it left to right, top to bottom. You try and then you figure it out, and then it might take you years when you eventually figure it out, right?
And then, and NLM is a pair, is a person that you kind of collaborate with, and you work with. But the software is being created, like there’s a saying, the best ideas, the best idea should win, right? But I find that the best idea is not a particular one idea.
The best idea is a Frankenstein of all the ideas that you had in the room that you break down and then you, sometimes you make a completely different thing it’s not like, you know, with, with people you, you cannot build it, but with LLM it will be like this perfect thing that will build exactly what you want. Very often you don’t know what you want. You have a feeling. Or like Steve Jobs was saying.
Keep on going. I’ll tell you when we’ll, we’ll reach there. What does it even mean? What kind of a product management device is that? Right? And then so, so there’s a lot of ideas and a lot of directions, but you have to evaluate them, verify them and think about them as a team. And LLM is just a member of that team.
Matt: I think there’s also something interesting about how, and this, I think in a number of industries, the idea of automating out the kind of the grunt work is.
, attractive because you can take out human cost if you can get the, the LLMs to be able to generate stuff that otherwise would be fairly tedious to produce. But there is something about learning the art of a particular profession through doing the tedious stuff. It’s putting, putting the time in to be able to understand how the fundamentals work.
, how will developers learn to do good development? If a lot of the, drudge work is taken away, which is the sort of stuff that you may give to a more junior developer.
So who is starting to learn the. I don’t know if there’s an answer to that. I don’t know if it’s an unfair, , fear, but it does feel that the stuff that is most likely to get automated will be the stuff that actually does serve a value in the same way that, I don’t know. Meetings in person in offices served a value, but it wasn’t the meeting.
It was all the stuff that sat around it.
Yaroslav: I think, you know, LLMs just can produce much more things in a shorter period of time, not necessarily the right things. , there’s a lot of discussions, and there’s a lot of speculation. There’s some principles, , like if you put a lot of compute into, a task and generate a thousand ideas and then validate them and then build on top of that, you can actually like for like a million dollar, you can make, uh, allow them to do amazing things as compared to just sending one prompt, which is not well, you know, , documented.
But this is the same with humans. , I tend to look at an LLM and then work with him and use the guidance of an LLM like I would be of a wise friend who doesn’t judge you, who generally has genuine knowledge on stuff, who is hard working and is not trying to just like, you know, look other way when you have a conversation with him, and he doesn’t…
Talk shit behind your back on you and then basically when you do that, and then you would you would you know Sometimes people say well, I told the left to write a book for me and it didn’t I was like, well Did you try it with a human to write a book for you? Then like you you have to show some respect you have to explain some certain things You need to explain the context and what’s happening and then where where you’re at Where are you struggling business like just you know, treat them treat treat the thing with respect I think, conscious approach is the center of all of it.
So if you consciously and genuinely are trying to do something,, with people or with LLMs, and then you really, consciously means you need to understand. , but the world will probably divide into people who genuinely want to do something and then LLMs will help them with that and the world will be just watching an infinite amount of Netflix generated by the LLMs in some kind of a virtual reality matrix.
Matt: Another theme for, , your business is you are very much a distributed hybrid global thing. And part of that is because you’ve had to be able to think, I guess, quite Uh, well, you’ve had to deliver quite a lot of change to how you operate over the last two or three years, because your business started based in Ukraine and you’re now working in all sorts of places across the world.
What was the last few years been like be able to have to re rethink how you operate?
Yaroslav: Actually, we, we, we think how we operate 10 years ago, when we allowed people to do remote work before it was cool. Way before COVID and people were, you know, traveling to Thailand or Bali before that was cool and, and, and spending some time there for months and working from there and just generally synchronizing their hours into around European hours.
It doesn’t matter if you’re like in America, like I get up at 4am to work with people. Or when you’re in Bali, let’s say you, you were up, up until midnight. And, and that’s the problem with the person that, that moved, that shifted. So you live in the morning and then, you know, you work at the night or in US, it’s the other way around.
I started very early. And then after 2 p. m. I basically have my, my life to, to attend to. , and we’re in 24 countries right now and in 79 cities, which is interesting. I’m sure his numbers, right? But it might be, there’s a lot of solo of people in just one city, right? We, because of this consciousness approach and because of this artistic beginning, humble artistic beginnings, we always were hiring people who are able to participate in this creative process and actually deliver products.
, And then we created this core, , principles and we saw what works, what doesn’t work. And, , we’re now, we hire about 0. 66 percent out of the market. And in the last three years, we reviewed 50, 000 candidates out of them, we hired 150. people. So it’s, , it’s a ridiculous amount. , I use coupler, our data product, just to pull the calendar data, calendar lead data and calendar data.
And we have, , 160 meetings in the past three weeks with candidates. Is it an intrapol or it’s a thing called the parent interview, or that was a full day where we spend, , time with the person or half of a day. It’s just a lot of, a lot of selection process. We, we really put a lot of efforts into selecting people and, , that’s why it works.
And we don’t care where you’re at. We just care if you are able to be this part of this process and then there’s enough people to onboard you. Um, you know, there’s enough people with 10 years of experience and then nine years of experience, eight, seven, three, two, one, and then we have been growing. Uh, we didn’t grow for most of the time we started growing.
And before that, we were pretty static, you know, we’re at the same size at around the same revenue for like 10 years. And then in the last six years, we started growing like 20, 30, 40, 60, 60 percent per year. Um, and, um, just using all those principles that we learned and everything that we practiced and, and.
You know, sometimes we were, there was a time when I was helping a lot of people and spending with them and holding them and trying to help them to deliver something. And then I looked around that I’m helping them to deliver like regular daily job. And the rest of the people are delivering this daily job by themselves with zero help.
I was like, wait, wait a minute. It’s not fair to those people maybe I help them to deliver the job better rather than how those people deliver their basic job And this is what’s our one of our principles. Is that a person has to be good at basic Work and if they’re not a good basic work, we cannot really help them because there’s a lot of help needed For that regard, and there’s, there’s no, you know, there’s no age, uh there’s, there’s no age or race or whatever that is a part of this requirement.
It’s just, can you do a basic job? And then, you know, and sometimes people complain, they’re like, you didn’t give me a great enough description yet. Maybe you didn’t give a perfect description, but those other 150 people are productive with this description and you’re not. And I’m sorry, but we cannot produce this description.
You have to be more dynamic in this. , you know, like just, just learn this approach and learn it for your peers. And, and then if you can, great. And if you cannot, then, then we will just have to move on to someone else out of the one of the 50, 000 people that we look through. And the other people are very, you know, they’re happy.
They’re, they’re productive. They, they enjoy it. They love the principles. They come to do it.
Matt: How many of the people that go through that selection process are then not a fit for you? So if you talk about that kind of the people who can’t do the basics there
Yaroslav: I would say people always can do the basics especially for the injury sake or for the first three months.
, But then they have their own Ways to look at things and then they they might kind of degrade their quality of work on a daily basis And say this is too much for me But we’re always very transparent and that’s why we have a full day and we were super transparent of what is it that we expect.
It’s more like 20 to 30 percent.
Matt: And in terms of being able to maintain a culture within all of this, you talked there about onboarding and having people who are able to support others coming into, what other things do you do to be able to establish some sorts of identity as an organization, when you’ve got such a distributed workforce?
Yaroslav: You know, there are distributed, but they’re still working a particular team, , on a particular thing. And we always make sure that we never start a project with a new person, just a new person. That’s it. And there’s no, like, even management oversight is too little. , You have to be a part of some group of, there’s so many factors that get slashed throughout daily work.
And we always make sure that there’s someone who understands our culture, who can pair. The person, I mean, if you pair the person for, you know, five days a week, uh, for a particular period of time, you, you exchange. Everything., thoughts, culture, the principles, comments on your code. There’s so much that happens.
And this pairing is very, very important for all, for all sorts of work. , and we do it, you know, in accounting, in recruiting, in support, in anything., and then this culture is kind of shared. Actually, McDonald’s is pretty good at it. Their, their pair teach their people. Um, you, you would come in and you would see the junior person and the other person explaining them and showing them, learn by doing, basically, , what is it that they’re doing.
And then this culture is being shared. Now this culture is, , , , everyone is a part of the culture and they were executed within the culture. I have no immunity. You know, if, if I break the culture, people can, can bash me and do whatever they want to me. So I’m, I’m, um, I’m the same member of the, of, of the s culture.
So, and then, you know, and then this culture gives birth to, to really great things and I think it’s, it’s an, uh, organism on its own basically nowadays.
Chris: Ah, thank you . , that’s a really interesting, uh, story, uh, around your business and the culture within, , what have you got planned for this week in the, in the week ahead? Is it an exciting week? Have you got plenty going on?
Yaroslav: It’s a very exciting week. , we have some product launches that I’m looking forward for.
They’re being in work for quite a while and when it’s a product that we launch in addition to our current product, we really try to make sure that our clients will get a great experience right on. So we’re going to deliver our second product to the Atlassian marketplace and third, I think, at the same time.
And that’s an AI product and AI products are very exciting, , nowadays. And then there’s, there, there’s some deadlines, some interesting deadlines that we need to meet, like, uh, the Atlassian event is coming out in Las Vegas in April and the deadline to submit a speech. To it is, uh, 1st of December that we just figured yesterday that it is 1st of December, so, uh, we need to, put it together.
What are you doing? What are you
Chris: doing here? You should be writing a speech.
Yaroslav: Yeah, I am, I am. Actually, I’ll just use the whole podcast as a speech. Ask the, the LLM to, to make a presentation out of it and I think it’ll work. And then there’s, there’s a lot of family stuff. I have four kids. I have four daughters.
Two year old, six year old, 14 year old, and 20 year old. And then each one of them always has a big drama and multiple medium dramas and a couple of small dramas. You know, after, after I seen 50, 000 people in recruiting, I was like, well, I need to make my daughters be great and then go try to figure that out.
Right. And we did like, I don’t know, 50 products, probably 50 software projects throughout our life cycle. And now I’m only starting to figure it out. So if I would have 50 kids, that would make sense. I would know how to raise them, but you never know how to raise them. And, uh, I think we’re getting a dog soon as well, so we need to figure that out.
So, that’s it. Every week, very fun week. Nowadays, life is very slow. One day is so much, is a lot. Like, it can change, it can make it or break it. So it’s very reasonable.
Chris: All right, you know, I don’t envy you, , four kids and a dog coming up. But, , then again, you could be, , Matt. What’s going on? What’s going on in your life, Matt?
What’s your, week hold?
Matt: , it’s going to be a relatively quiet week because a lot of my colleagues are up in Manchester for a series of events, , which I’m not going to be there for. A series of unfortunate events. No, no, no. It’s unfortunate I will not be at the events, but there will not be unfortunate events.
It’s all good. And then, , the weekend will beckon to the annual ceremony of buying the Christmas tree, which involves initial excitement and, , a sense of festive cheer. And then by tree number 76, which isn’t up to the standard that, uh, the, the matriarch of the house insists upon for symmetricality of trees, even though they are naturally growing objects that are not symmetrical.
, and then tempers afraid, shouting happens, and eventually we go, Oh, that one will do! And, uh, it gets shoved into the back of the car, and then we take it home to stick it in the corner of the house and… Stick stuff on it. , i’m looking forward to it even though and we were laughing about the fact that it almost invariably Ends up in in arguments and short tempers because the amount of decision making that has to be involved in the tree How about you?
How are you? What’s your week head looking like?
Chris: Well, actually, yeah, quite quiet compared to last week. , not so much travel. , but the house has been upside down for weeks because we’ve had the kitchen and I’ve been decorating all weekend. , something I enjoy doing and then no doubt it will be Christmas decoration time, but we, we have the annual, , tradition of getting me getting the tree out of the loft.
That’s, we have a, we have a, a You know, faux tree that, gets put up every year. It’s a perfectly good tree, and I, and I have done the real tree thing. And you think I’m dragging one of them damn things into my house, only to have to drag the damn thing out again after Christmas, shedding needles all over the place.
No, no, I’m not doing that. So, , the, , plastic one will do.
Matt: Oh, and who says the spirit of Christmas is dead? There we go. Anyway, , that’s it for another week. We have two shows now left until the, , The Christmas break. So, next week we have got a look ahead in a special edition of Ask WB40 to ask about things that might be happening in 2024.
So, , drop us a line if you have some ideas for that. And then we’ve got one more guest before the end of the year and then we’ll be taking a break over the Christmas festivities. , Yaroslav, thanks again for joining this week. It’s been a delight. Thank you. And we will be back, same place, next week.
Yaroslav: Thank you for listening for WB40. You can find us on the internet at wb40podcast. com and all the good podcast platforms.