AI & I

Alice Albrecht, a cognitive neuroscience Ph.D. from Yale and Senior Director of AI Product at SmartNews, dives into how AI is revolutionizing science. She discusses the importance of large-scale data gathering and the shift towards personalized, contextual predictions in research. The conversation explores the ethical challenges of AI in mental health, particularly in treating OCD, and highlights the transformative power of data in understanding psychological dynamics. Albrecht also anticipates a new era for scientific publishing through greater transparency and open data sharing.

Chris Pedregal is the cofounder and CEO of Granola, an innovative AI-powered notepad designed to enhance meeting productivity. He shares insights into his product development philosophy, emphasizing the blend of intuition and user feedback for creating meaningful products. The discussion also touches on the importance of emotional health in startups, the transformative power of travel for creativity, and the challenges faced by AI entrepreneurs in a competitive market. Chris's passion for building AI tools with 'soul' shines through in this engaging conversation.

Steve Krouse, Cofounder and CEO of Val Town, discusses the impressive evolution of AI in coding, highlighting how tasks once taking an hour can now be completed in seconds. He shares his journey into programming and the vision behind Val Town, aimed at empowering sophisticated developers while also appealing to non-coders. The conversation covers the transformative potential of AI tools like Townie, which simplifies app creation and enhances productivity. Krouse emphasizes the importance of clarity and understanding in the fast-paced world of AI.

Over the last few months at Every, we’ve launched two new AI products with tens of thousands of users, and we’ll release a third one before the end of the year. The weird thing is: We’re a media company with less than 10 full-time employees, and we’re mostly bootstrapped.That’s not how things are supposed to work in startups.When we were first starting Every Studio six months ago, we were told a million reasons why it wouldn’t work: divided focus, not enough money, and the biggest one—it would be too hard to find talented people to run the products we build.Yesterday, we proved out one of the biggest risks to this strategy: We launched a brand-new version of our AI product Spiral with Danny Aziz as its general manager. Danny left a $200,000-a-year salary to come chase his dreams with us. So we decided to take this moment to pull back the curtain and ask, Why? Why did he join us? And why is the model we’ve built working so far? What have we learned about what happens when you mix media, software, and AI in a single organization?That’s why I invited Danny and Brandon Gell, Every’s head of Studio, on the show. We get into the details of Every’s business model, how new technology and the right people can make the flywheel turn, the power of learning by doing and building from real needs, and where each of us sees ourselves one year from now. Here is a link to the episode transcript.This is a must-watch for anyone who wants to build a business on their own terms, and have a lot of fun while doing it. If you found this episode interesting, please like, subscribe, comment, and share! Want even more?Sign up for Every to unlock our ultimate guide to prompting ChatGPT. It’s usually only for paying subscribers, but you can get it here for free.To hear more from Dan Shipper:Subscribe to Every: https://every.to/subscribe Follow him on X: https://twitter.com/danshipper Links to resources mentioned in the episode:Danny Aziz: @DannyAziz97Brandon Gell: @bran_don_gellTry Spiral here: https://spiral.computer/  More about Every’s product incubation arm: https://every.to/p/introducing-every-studio 

Everyone told Vicente Silveira that his startup—a GPT wrapper—would fail. Instead, one year later, it’s thriving—with about 500,000 registered users, nearly 3,000 paying subscribers, and over 2 million conversations in the GPT store. Vicente is the cofounder and CEO of AI PDF, a tool to help you summarize, chat with, and organize your PDF files. When OpenAI allowed users to upload documents to ChatGPT, the consensus was that his startup, and all the other GPT wrappers out there, were toast. Even when some of his competitors closed up shop, Vicente believed they could still create value for users as a specialized tool. The AI PDF team kept building. Today, AI PDF is one of the most popular AI-powered PDF readers in the world—and they did it with a five-person team and a friends-and-family funding round.I sat down with Vicente to understand, in granular detail, the success of AI PDF.Vicente explains how staying small and specialized is a key strategic advantage for his business. We get into why lean startups are better positioned than companies like OpenAI and Anthropic to create cutting-edge solutions for users, the role early adopters of technology play in shaping the market for new products, Vicente’s candid take on raising capital as a growing startup, and his thoughts on the emerging role of AI managers who will be responsible for overseeing AI agents. We demo an agent integrated into AI PDF, prompting it to analyze a bunch of recent articles from my column Chain of Thought and write a bulleted list of the core thesis statements—and even pit AI PDF against Perplexity live on the show. This is a must-watch for small teams building profitable companies at the bleeding edge of AI. If you found this episode interesting, please like, subscribe, comment, and share! Want even more?Sign up for Every to unlock our ultimate guide to prompting ChatGPT. It’s usually only for paying subscribers, but you can get it here for free.To hear more from Dan Shipper:Subscribe to Every: https://every.to/subscribe Follow him on X: https://twitter.com/danshipper Links to resources mentioned in the episode:Vicente Silveira: @vicentesAI PDF: https://myaidrive.com/  Dan’s piece on the allocation economy: https://every.to/chain-of-thought/the-knowledge-economy-is-over-welcome-to-the-allocation-economy    

Prompt engineering isn’t just about telling AI to solve your problems—it’s about knowing which ones to solve. Yet there’s a mismatch between the people who can identify the right problems—experts with deep domain knowledge—and the technical infrastructure required for developing and refining prompts. Jared Zoneraich, the cofounder and CEO of prompt engineering platform PromptLayer, is bridging the gap with a platform on which non-technical experts can manage, deploy, and evaluate prompts quickly.The role of human prompt engineers, however, has been the topic of controversy, with some arguing that AI can optimize prompts better than us, while others suggest that more capable LLMs eliminate the need for meticulously crafted prompts altogether. I spent an hour talking to Jared about why he believes prompt engineering isn’t becoming obsolete. He also tells me everything he’s learned about writing a good prompt and what the future of AI tools looks like. Here is a link to the episode transcript.This is a must-watch for prompt engineers, people interested in building with AI systems, or anyone who wants to generate predictably good responses from LLMs.If you found this episode interesting, please like, subscribe, comment, and share! Want even more?Sign up for Every to unlock our ultimate guide to prompting ChatGPT. It’s usually only for paying subscribers, but you can get it here for free.To hear more from Dan Shipper:Subscribe to Every: https://every.to/subscribe Follow him on X: https://twitter.com/danshipper Links to resources mentioned in the episode:Jared Zoneraich: @imjaredzPromptLayer: @promptlayer, https://www.promptlayer.com/A couple of Steven Wolfram’s articles on ChatGPT: What Is ChatGPT Doing … and Why Does It Work?, ChatGPT Gets Its “Wolfram Superpowers”!   

Notion cofounder Simon Last told me everything he’s learned from integrating an AI application into a platform that has over 100 million users.Simon likes to keep a low profile, even though he’s the driving force behind Notion AI, one of the most widely scaled AI applications in the world.In this episode, we get into how AI changes the way he builds software since the days he cofounded Notion with Ivan Zhao in 2013. He talks about the challenges that arise because AI doesn’t follow the deterministic rules of traditional software, and how he designs evals to build AI systems that are reliable at scale. Simon tells me about the AI tools he uses to code and how he would think about rebuilding Notion from scratch with them. He also shares his thoughts on how the growing capabilities of AI are redefining human roles, and argues that we have the responsibility to shape technology to align with our collective vision of the future. This is a must-watch for anyone interested in building reliable AI products at scale.If you found this episode interesting, please like, subscribe, comment, and share! Want even more?Sign up for Every to unlock our ultimate guide to prompting ChatGPT. It’s usually only for paying subscribers, but you can get it here for free.To hear more from Dan Shipper:Subscribe to Every: https://every.to/subscribe Follow him on X: https://twitter.com/danshipper Links to resources mentioned in the episode:Simon Last: @simonlastNotion AI: https://www.notion.so/product/ai The AI code editor Simon uses: CursorOpenAI’s definition of AGI that Simon ascribes to: https://openai.com/charter/ 

In this discussion, Matt Cynamon, an innovative force at Union Square Ventures, reveals how they’re crafting an AI operating system to harness the firm's collective wisdom. He shares insights into AI tools like a chatbot trained on thousands of articles, portfolio trackers, and a unique meeting notes tool. They also explore a fascinating art project, The Dream Machine, that generates visuals from conversations. Matt's engaging anecdotes showcase how AI enhances creativity and collaboration in venture capital.

Yohei Nakajima⁠ leads a double life. By day, he’s a general partner of a small venture firm, ⁠Untapped Capital⁠. By night, he’s one of the most prolific internet tinkerers in AI. (He also sometimes works on ⁠automating his job as a venture capitalist.)He’s the creator of BabyAGI (@babyAGI_), the first open-source autonomous agent that went viral in March 2023. Yohei has since released seven iterations of BabyAGI, a coding agent called Ditto, a framework for building autonomous agents, and, most recently, BabyAGI 2o, a self-building autonomous agent.I sat down with Yohei to talk about:What feeds Yohei’s drive to create new toolsThe evolution of BabyAGI into a more powerful version of itself What Yohei learned about himself by tinkering on the internetYohei’s personal philosophy about how the tools we build our extensions of ourselvesWhy founders in AI should think about their products from a modular lens, by addressing immediate problems while enabling growth in the futureYohei’s insight into a future where models will train themselves as you use themWe experiment with Ditto live on the show, using the tool to build a game of Snake and a handy scheduling app. Yohei also screenshares a demo of BabyAGI 2o in action.If you found this episode interesting, please like, subscribe, comment, and share! Want even more?Sign up for Every to unlock our ultimate guide to prompting ChatGPT here: ⁠https://every.ck.page/ultimate-guide-to-prompting-chatgpt⁠. It’s usually only for paying subscribers, but you can get it here for free.To hear more from Dan Shipper:Subscribe to Every: ⁠https://every.to/subscribe⁠ Follow him on X: ⁠https://twitter.com/danshipper⁠ Timestamps:Introduction: 00:00:59BabyAGI and its evolution into a more powerful tool: 00:02:26How better models are changing the way Yohei builds: 00:05:00Using code building agent Ditto to build a game of Snake: 00:08:10The ins and outs of how Ditto works: 00:13:24How Yohei gets a lot done in little time: 00:19:21Yohei’s personal philosophy around building AI tools: 00:21:50How Yohei experiments with AI as a tech-forward parent: 00:33:13 Demo of Yohei’s latest release, BabyAGI 2o: 00:39:29 Yohei’s insights on the future of AI tooling: 00:51:24Links to resources mentioned in the episode: Yohei Nakajima: @yoheinakajima, ⁠http://yohei.me⁠ Untapped Capital: @UntappedVC, ⁠https://www.untapped.vc/⁠ My first interview with Yohei, around the time he released BabyAGI: ⁠https://every.to/chain-of-thought/this-vc-is-slowly-automating-their-job⁠ The other AI tools Yohei has created: ⁠Ditto⁠, BabyAGI 2, BabyAGI 2oThe tweet thread about AI bots being let loose on a Discord server: https://x.com/AISafetyMemes/status/1847312782049333701 

Simon Eskildsen is a learning machine. I first interviewed him in 2020 about how he leveled up from being an intern at Shopify to becoming the company’s director of production engineering by reading and applying insights from hundreds of books.A lot has changed over the last four years. Language models have made it possible to access and contextualize information faster, easier, and more cheaply than ever before—and in this episode, Simon and I talk about how this changes the way he learns. Simon is now the cofounder and CEO of AI startup turbopuffer, which is building a search engine that makes vector search—an approach to information retrieval that uses machine learning to gather context—easy and affordable to run at scale.We spent an hour talking about how he leverages LLMs’ contextual intelligence to supercharge his learning, such as helping him pick up new words as a non-native English speaker, do odd jobs to maintain his rural cabin in Quebec, and articulate technical concepts in legalese. As we talk, we screenshare through his Anki setup, including the flashcard template he finds most useful, and the custom AI commands he’s created in productivity software Raycast. Simon tells me about the clutch of AI tools he experiments with for journaling, writing, and coding, as well as his thoughts on how language models will fundamentally reshape the way we learn. Here’s a link to the transcript of this episode.  This is a must-watch for note-taking aficionados and anyone who wants to supercharge their learning with AI.If you found this episode interesting, please like, subscribe, comment, and share! Want even more?Sign up for Every to unlock our ultimate guide to prompting ChatGPT. It’s usually only for paying subscribers, but you can get it here for free.To hear more from Dan Shipper:Subscribe to Every: https://every.to/subscribe Follow him on X: https://twitter.com/danshipper Links to resources mentioned in the episode:Simon Eskildsen: @SirupsenSimon’s startup, turbopuffer: turberpuffer.com, @turbopufferMy first interview with Simon in 2020: https://every.to/superorganizers/how-to-build-a-learning-machine-299655 The productivity tool through which Simon uses LLMs, Raycast: https://www.raycast.com/ The other AI tools that Simon is experimenting with: voice-to-text tool superwhisper, copilot for developers Supermaven, code editor Cursor