Podcast appearances and mentions of Paul Graham

#743: Everyone talks about a loneliness epidemic — but Luke Burgis argues the opposite problem is just as real: we now have so much easy, frictionless community that we never have to develop a solid sense of who we actually are. Luke Burgis is a professor of business at The Catholic University of America and the founder of the Cluny Institute, and the bestselling author of Wanting: The Power of Mimetic Desire in Everyday Life. He returns to the show to talk about his new book, The One and the Ninety-Nine. In this episode, we discuss: Why having too much easy community can be just as damaging as having none at all How to tell whether your beliefs are actually yours — or something you inherited without ever examining it Why cutting people off has quietly become the default response to conflict How to stop shrinking yourself just to keep other people comfortable The zero-tolerance rule Luke enforces at his own company to kill passive-aggressiveness before it starts How Luke actually decides who to trust, hire, and build relationships with What an existential crisis at 29 — and five humbling years training for the priesthood — taught a successful entrepreneur about identity This episode is for anyone who feels the pull between fitting in and standing out — in your family, your workplace, or your online life. Luke offers a way to build an identity solid enough to hold up under pressure, without giving up on real community. ⏱️ TIMESTAMPS Note: Timestamps may vary slightly depending on dynamic ad placements. (7:42) Two competing drives wired into every human being (10:11) Why group pressure breaks some people and not others (17:39) Why cutting people off replaced working through conflict (19:39) The real number of close friends you actually need (26:51) Where your beliefs actually came from (32:00) You're not responsible for how someone else feels (36:09) The zero-tolerance workplace rule against passive-aggressiveness (41:22) How to actually tell who you can trust (53:33) The existential crisis that sent an entrepreneur toward the seminary (55:39) The humbling lesson hidden in a vending machine

https://youtu.be/7yugccgcgs8 Justin Nassiri, Founder and CEO of Executive Presence, is driven by the power of human connection and a mission to harvest CEO stories that reveal authentic experiences and valuable insights. By serving as thought partners to C-suite executives, Justin and his team transform personal perspectives, mistakes, and lessons into compelling LinkedIn content that builds trust and distinguishes leaders from generic, AI-generated voices. In this conversation, Justin introduces The Content Strategy Framework—Use the 40:30:20:10 Content Formula, Harvest Stories, Borrow Thought Patterns, and Apply Curiosity. He explains why leaders should build visibility through personal profiles, how skilled interviewers uncover stories executives may overlook, and why Thought Leader Ads can extend the reach of proven content. Justin also discusses growing through referrals and warm relationships, using 10-week improvement cycles to revisit every business process, and developing autonomous team members who use AI to solve problems while preserving human connection. — Harvest CEO Stories with Justin Nassiri  Good day, listeners. Steve Preda here, and my guest today again is Justin Nassiri, the Founder and CEO of Executive Presence, a fully managed LinkedIn thought leadership service for C-suite executives at growth-stage B2B companies. Justin, welcome back to the show.  Great to be back. Thanks, Steve.  So we just reminisced that it was three years almost to the day that you came here, and I can’t believe it. It feels like yesterday. But your business has grown dramatically during that time, so I think you have some new insights that you’ll be able to share with us, I’m sure.  Thank you. It’s good to be back. We just made the Inc. 5000 list, which is a first for me. I’ve never been on that before, but we’re celebrating that.  Yeah. Congratulations. That’s a great milestone to hit. And you only started in 2022, right?  Yeah, yeah.  So, pretty freshly minted, fast-growing Inc. 5000. So my question to you is, what is your personal why, and how are you manifesting it in Executive Presence, in your business?  There’s a very potent macro or micro why for me, and then probably a broader why. I think that the biggest why for me right now, like many parents, is my kids. I’ve got a three- and a seven-year-old. And so when I think of my professional life, I certainly think about not just providing for them but also trying to set an example of someone who is trying to do their best and trying to stretch and trying to grow.Share on X  And so I think that’s probably the highest leverage that I have. But I think the through line in the companies that I’ve done is just the value of human connection. My first company was all about companies using Instagram in a way to be more authentic and more genuine with their community, and that’s very true to what we’re doing now at Executive Presence, just really helping people connect to other people.  In this case, it’s executives using LinkedIn to connect with a broader audience, but I really feel like that human connection is so important, and I think it’s becoming even more important in the era of AI. So that’s a little bit more specific why in what I do right now.  Yeah, it’s fascinating how human connection is evolving in the age of AI, and I agree. I mean, I see that because there’s so much more noise out there, human connection is perhaps more important than ever. People want to make sure that they are talking to authentic people and hearing from authentic people who have authentic lived experience. So how does that impact communication for executives on LinkedIn? How do they have to evolve their voice or how they approach things?  I think if you look at LinkedIn in particular right now, I think it still remains the place where the largest source of our professional network is. And so I think there’s still a lot of value there, and I think that LinkedIn is facing a lot of growing pains. Specifically, I think that there are three things driving it. One is more people are just showing up on the platform. The secret is out, and people realize that there’s value to LinkedIn.  So, all things else equal, more people are more active on LinkedIn, which generates a lot of noise. And then the second thing is AI is making it easier to create content, so I think it’s creating not just more content, but lower-quality content. And then the third thing is it does seem as if LinkedIn is following what Facebook did over a decade ago and saying, “Look, to get reach, you’ve got to put money into ads now.”  And we saw that transition with Facebook company pages a long time ago, but it really does feel like you cannot get as much visibility today as you could have two years ago unless there’s some sort of ad buy behind it. And so I think those are kind of the three problems. I think that the answer is still there’s value in showing up, but I think that you have to show up even more human. I think that the experiences that make you unique and the mistakes that make you who you are and the things that you know and the stories that you can tell, those still hold value, and that differentiates you from generic or AI-generated content.  I also think that there’s value to using LinkedIn ads, and we can talk about that, but I think that that has to be part of one’s strategy now if you’re trying to significantly use LinkedIn for what it’s good for, which is brand building.  Yeah, and maybe this is a slight question, then we’re going to talk to you about the framework. But as I understand, LinkedIn now allows individuals to also boost their posts as opposed to just companies, which used to be the case in the past. So how does it impact companies? Is there more emphasis now? Is emphasis shifting to individual posts because there’s no real reason to build up the company pages? How is that evolving?  Yeah, I mean, four years ago when I started the company, before these Thought Leader Ads, these individual people ads were a thing, I still would’ve said to you, “Look, there’s not much value in company pages. People connect with other people. They’re not going to connect with a faceless organization.” So even four years ago, I would’ve said, “Man, if you really want to raise visibility for your organization, you’ve got to do that through your key leaders.  You’ve got to do that through actual names and faces and voices and perspectives.” And I think that the ads make that even more pronounced now because I can now take my personal post as Justin Nassiri. If it does well on LinkedIn, I can put a $50 or $500 ad buy against it, and I can make sure that essentially specific people are going to see my post. If I’m selling to CEOs of tech companies in Cincinnati that have grown 10% last year and they’ve been at their company for five years, I can have an insane level of targeting, and I could put my personal content in front of them in a way that most people still don’t realize is an ad.  It will say, “Promoted by,” and then the company name, so “Promoted by Executive Presence.” Most people, when scrolling, don’t even notice that. Yeah. And so I think there’s a tremendous opportunity then to take very human and personal content and put it in front of exactly whoever you’re trying to get in front of.  Yeah, I love that. You also say on your LinkedIn page that it takes an executive only 90 minutes a month to actually work with you guys and have you amplify them. So how do you extract all those personal stories and experiences? What is your framework for that so that when you meet with an executive, you’re able to create those posts without them having to be involved? Yeah, I think the framework is pretty universal for anyone listening. We tend to start with the content strategy. And the way that we typically start is we’ll say, okay, 40% of the content we would call industry thought leadership. And that is great if you can talk about current events and relate them to your industry. It’s great if you can share things about your industry that no one knows or that you disagree with people.  But do that 40%, which is the biggest of any of the categories, as really the education, as the subject matter expert. You are showing up as an authority in cybersecurity, or you are showing up as an authority in leadership, or you’re showing up as an authority in B2B supply chain. And I think the key here is the more niche, the better. The more narrow, the better. We are not Joe Rogan. We’re not trying to get 300 million people to look at your content. We want to get in front of a very narrow group of people, typically prospects, customers, potential employees, potential investors.  We really want to narrow where your voice can be fairly large in a very finite realm. So that's the industry thought leadership piece.Share on X The second one, about 30% of the content, we call it leadership and career journey. And what we’re trying to do here is a blend of humanizing the executive while also giving them credibility. And so if we are working with someone who is a CEO at a company, well, they’ve done things prior to that. So what did they learn in college or a previous workplace? What was a mistake that they made? What was a mentor that said something to them?  So that’s a way of us imparting one of their values or something that they know, but wrapped in a story from their history, which humanizes them. And that would be generally 30% of the content. Twenty percent, obviously they’re doing this to promote their company, so 20% would be about their company, spotlighting an employee, recent events, things like that. Again, trying to do it through stories if possible. And then the last 10% is usually the highest-performing 10%, and we would call that work-adjacent content. So we want to, again, make them a three-dimensional person. What do they do outside of the office? Is that family?  And we usually use what we call the dinner party test for this. If you were with prospects and potential employees, what’s fair game to talk about over dinner and drinks? Some people would definitely talk about their kids. Some people would never talk about their kids. Some would talk about their hobbies. Some would never talk about that. So that's a good filter to figure out what they could talk about that's not just always talking shop. So that's kind of the framework that we use.Share on X  But I think that the way that we harvest this information is really the skill of the people that I hire on my team. It’s people who are really good at pulling insights out of someone and getting someone to open up and having that heat-seeking missile approach of, what is a story that they’re sitting on that they don’t even realize is a compelling story? And that’s one thing that has stood out. Some of the best-performing LinkedIn posts, the person didn’t even think that that would be interesting to anyone else.  We’re often not the best filter for ourselves of what’s going to land. And that’s one of the values of LinkedIn, is that you can actually put out ideas and stories and insights, and very quickly, in an 18-hour time period, get signal from the market if people value that from you or not, and then follow that trend.  Yeah. That’s fascinating, and it sounds a little bit like being a ghostwriter for someone, that you can really get those stories out and you can have them open up so that their brain is going to surface those things that maybe they don’t think about. Maybe they are not extroverted and they won’t be able to bring this up on their own. But if you catalyze it, then they come to life that way.  Yeah. I think of it as a thought partner. I kind of realized this because I hosted a podcast for a long time as well, and you kind of realize the power—exactly what you’re doing—the power of curiosity and the power of distance, right? You are showing up, you’re curious about my experience.  You have enough distance from it that you’re asking questions that might even seem intuitive to me, or it might be one of those things where I’m like, “Well, everyone knows this.” But then you bring an outsider in, and it’s like, “No, not everyone knows this,” or, “I think people would find this interesting.” So having that thought partner to be the outside observer of what others would benefit from.  Yeah. That’s amazing. That’s a real skill, and that also brings in that human skill that an AI is not going to be able to prompt those kinds of questions, that kind of curiosity, that there’s an emotional driver behind it. That’s a very journalistic trait, I suppose. That’s a new form of journalism that you’re practicing here, isn’t it?  Yeah, it is. And the type of person I hire to do that are ex-consultants because they’re really good. They’re really good at coming into a business and understanding the objective and understanding how to get very senior people—because we work with CEOs of publicly traded companies and CEOs of smaller companies—to really get very prominent people to open up. And being comfortable interrupting or redirecting or pushing back, it really is a unique skill set.  Yeah, I love that. So that brings me to my next question. What drives growth in your business? How did you get, in four years, on the Inc. 5000? What was the engine here, the fuel?  Well, it’s so funny because I ask this of every CEO I meet with as well, to learn from them. Everyone always says referrals, so I’ll be generic and say referrals do drive—it’s probably the single biggest source of revenue. When I first started the company, I actually used LinkedIn. And so this actually came out of another company that I was running, and they had the idea. And so I went to LinkedIn and I said, “I think that the type of person who would be interested in this is CEOs of companies with at least 50 employees.” And I put that into Sales Navigator, and it came up with a couple hundred first-degree connections.  And I just sent out a fairly generic message of, “Hey, Steve, just wanted to give you a quick update. I’m launching this new service, and this is what we do. Let me know if you know of anyone who would like to chat.” And I actually got probably 60 or 70K in monthly recurring revenue from that, of people I wasn’t really aware of what they were up to. One of the guys who’s still a client, I had met with him 10 years previously when he was an investor, and then now he was CEO and founder of a company that ultimately went public.  So I wouldn’t have thought of that person, but that’s a great thing about LinkedIn, of saying, “Here’s someone who might be interested in what I’m doing.” And so that sort of outbound of warm connections has played a role. Obviously, I’m active on LinkedIn. I get a lot of leads from that. I do a fairly good job of keeping in touch with my network and seeing when people might be needing us. I think, back to the human connection, I do think conferences are playing a bigger role.  I’m starting to go to more conferences and realizing the value of meeting people in person and how that kind of seems to accelerate the process of building trust and building relationships. We do cold email. We do AdWords. We do a newsletter. We do a lot of content marketing, and so I think each of them plays their own part. But referrals certainly are the biggest one. LinkedIn is probably number two. Yeah. That’s very interesting. And do you see a lot of competitors? Is this a crowded field?  It is. We have expanded beyond LinkedIn largely because of the competitive nature, where I think just as people have flocked to LinkedIn. I would say the biggest faction is a lot of solopreneurs. A lot of individuals will work with a few executives, and so that’s probably the lower-level competitors. There are a handful of companies that are doing something similar to us, and then more established PR companies that will say, “Yes, we do LinkedIn as well.” But I think at this point, we’ve got the deepest track record of executives.  We work with over 400 now, which, as far as I’m aware, is the largest set. We actually present our data to LinkedIn every year because it is the largest data set of just executives rather than influencers. And I think the play for most professionals and most executives is different than what an influencer would do on a platform like LinkedIn, and I think it’s important to do what’s appropriate for an executive.  So you mentioned on your LinkedIn page that you bootstrapped this company to three million ARR.  Yep.  So what do you expect to be different from going from 60,000 monthly recurring, 700 ARR, to three million, to going from three million to 10 million? How is it going to be different?  Yeah, that’s such a good question. And I balance this because I really like Paul Graham’s thought that you have to do things that are not scalable to be able to scale. And so oftentimes, I’m looking for things to do consistently, but it’s really helpful for me not to constrain myself in that way and to think of things that—it still feels like guerrilla warfare at times—what are little things that we can do to get an edge?  I think that the thing that I think about most right now in getting from three to 10 million is creating a machine for experimentation, and experimentation not just in our service, but also in our sales and marketing. And so how do we create a culture? Let’s just take the service side of things. I never want our service to plateau. I have run a company before where our product became stale and a competitor put us out of business. I never want that to happen again. And that the way that we minimize the probability that that happens is that we are always experimenting.  We are always listening to our clients and understanding what else we could do to make their life better.Share on X But then we’re also looking at the market and thinking, what else might our clients not even realize they need but would benefit from? And I love that phrase from Henry Ford, “If I had listened to my customers, I would have built a faster horse.” I think that there’s a value in listening to customers, but also a value in being one or two steps ahead of them. And so, for example, one of the things that we’re heavily looking at right now is the thought of AI visibility for executives.  And where GEO is getting more and more prominence for organizations, we see a world where it’s very important not just to cultivate a human audience and a human group of people who view and like your content and respect you, but also essentially cultivating an AI audience and making sure that your content is visible by LLMs and making sure that Claude and Perplexity see you as the authority and are referencing you. And so that’s something that our clients aren’t yet asking for, but we’re already developing a solution and a thesis because we think that that’s the way that the world is going.  But the central point is, how do we create an engine for experimentation so we are always testing out new things and seeing if they work, and reinvesting in the ones that work and letting go of the ones that don’t work? And I think that if we can do that for our clients and for ourselves, it always keeps us evolving. We’re always upping the game. We’re always improving, because I think the moment that we stop doing that, that’s when we stagnate or that’s when someone else comes along and puts us out of business.  So how do you maintain this alignment and this entrepreneurial energy? Because experimentation is innovation. It’s entrepreneurship in your business. How do you perpetuate it? So as you’re growing the business, you’ve got 40 people now, maybe you’re going to have more, you’re going to have AI agents running around. So how do you keep that experimentation and this entrepreneurial energy as you are getting further and further from the newest hires?  So the first thing that I love—and this is my VP of Ops, Shelby, who came up with this—but I really like it. She instituted a 10-week cycle composed of one- to two-week sprints. And she oversees all of our client work. So what she did is she broke everything we do for our clients into different sections. She instituted a system of saying, “Okay, these are all the different things we do. If you have an idea about how to do something better, or if you have a complaint about how something’s not working, or if a client mentions something, I want you to put it in this spreadsheet so we can keep track of how we can improve each thing.”  And what we’re going to do is, this week we’re doing a sprint on interview questions—how we prepare our interview questions for our clients. We are going to invest a week-long sprint in improving that process through technology, AI, and processes. And then we’re going to go through every other aspect. But guess what? Ten weeks later, we’re coming back to interview questions again. What that does, I think, in today’s landscape is not only does that keep us always thinking of improving something, but every 10 weeks we’re re-looking at specifically AI to see, what are its capabilities now? It’s changing so quickly.  From 10 weeks ago, it might be able to do something better or different. So let’s create a system so that we are periodically refreshing every aspect of our business from team feedback, client feedback, but also technological improvements. And it’s mind-boggling to think that that’s the pace now, and that 10 weeks might not even be sufficient in the future with the rate at which things are changing. That’s the best example I can think of how we’re trying to create that mindset of constant and incessant improvement.  And how do you build your team? Are you remote, or do you have an office somewhere?  We are all remote. Yeah, we are 100% remote. So we’re in seven different states, all still in the United States. It does pose challenges as you grow of how do you get in person and create connection. And so we’ll do things like happy hours online and different ways to get to know each other. But I also think, for those others listening who create a remote-first culture, you filter for people who thrive in that environment and do their best work when they have a fair amount of autonomy. And I think that autonomy for us has been helpful because we want people who are individual problem solvers, and I think that overlaps well with people who prefer a remote-first workplace.  Yeah, that’s fascinating. So if you had a magic wand and you could fix one thing in your business in the next 12 months, what would you do?  That is a really good question. I mean, it almost feels like the world is moving towards more entrepreneurs, not fewer ones, which I, for one, like. I’m a huge fan of both entrepreneurship and entrepreneurs. But if I could wave a magic wand, it would be making everyone on the team think like an entrepreneur, which would have probably been a liability 12 months ago. But the way I approach everything now is—I literally, right before this, was using Claude for my personal finances.  On the business side, it is now connected to QuickBooks, and on the personal side, I have a system of creating a report exactly the way I want it, which I, for one, love—the capability of AI to personalize things. I don’t have to use Mint or QuickBooks anymore. I can build it exactly in the way that works for my crazy brain. But I was updating and realizing that the tools have changed, and so, like, upgrading the way that I track my personal finances. And that’s just kind of very natural for me, and I do that for my kids’ menus, and I do that for everything. I kind of have a project or an approach on AI.  And I think that that’s a similar mindset that I need for my team, is thinking like an entrepreneur. How do you get better at everything? How do you get more efficient at everything? How do you look for new ways of solving things? I think it used to be that the entrepreneur was the visionary and setting the vision for the company, and everyone was more or less following orders or a system.  I still see a role for the visionary, but it's almost as if everyone on the team has to be their own visionary of envisioning how to improve their workflowShare on X and how to make themselves more productive and how to utilize tools and realize where that can make them more effective. And I don’t know that it’s going to be top-down anymore. I think that the advantage of the tools we have today is, like, the way that I use AI, Steve, might be completely different than the way that you use it. And that’s the beauty of it, is that our brains are different, our worldviews are different, our skill sets are different, but we can almost bolt this technology on us to make ourselves superhuman.  But the way that it works for you is going to be different than me, and that will be true in a team. So, very long answer, but if I could wave that magic wand, it would be imparting that entrepreneur mindset of finding problems and the best way to solve them, and never stop solving problems.Share on X  Yeah. So it’s just a feeling, and maybe I won’t articulate it well, but I’m just looking at your business. You’re building this remote team, and then you’re making everyone more autonomous so that they are working with AI to improve their productivity. In a way, it is supercharging everyone individually, but what about the team cohesion? So how do you make sure that people don’t get isolated?  Isaac Asimov has a novel which is in the distant future, on a distant planet, and basically everyone has 1,000 robots, and everyone is in their own world, and they just communicate on video screen because personal contact is no longer even appropriate. And everyone gets super rich and super efficient, but something gets lost.  So I wonder, how do you see the tension between empowering people, having AI help everyone be super productive in their individual way, in a remote culture? How do you keep this constellation together going forward?  Yeah. It still comes back to human connection for me. I think that, let’s just say on a client level, if my company’s doing our job well with our clients, if we are having them become thought leaders, having them grow influence and audience, I want that to lead to more human connection for them. And the way that I have seen that showing up is they go to a conference and people know them and people come up.  Like, people are literally—they might not have created connection before, but they recognize them, they know who they are, and that draws them to in-person interactions. That’s human connection coming out of what we do. For our team, I hope it leads to more human connection, that as we become more efficient, as we become better at what we do, as the company grows, we will get together in person more often. We'll be able to be in the room together and brainstorm because that becomes more valuable, and we're all craving that.Share on X  So, sci-fi is my favorite genre of literature and cinema, and I don’t think that we will ultimately end up with VR goggles on our head and not talking to each other. I think that in the same way that we have seen with social media, that it has a purpose, but we still want to be around each other and benefit from being in person. I think that we will become more and more like that. I don’t think it will drive us apart. I think it will lead to more connection. That’s my optimist view on it.  Yeah. Love it. Well, I hope you’re right.  Yeah.  Okay, so if someone who is a founder of a business, a growing business, or maybe a C-level in an enterprise, and they don’t have time to manage their LinkedIn, but they realize that they’re missing out with their thought leadership and they need help or they want to explore, where should they go and how can they connect with you and your colleagues?  Yeah, I appreciate that. I would just say, in general, every leader needs to know that their personal brand is going to impact both their career as well as their company, and them being able to articulate their viewpoints and what they think and believe is going to be a vital skill. And that could be on stage at a conference, it could be on YouTube, it could be on LinkedIn, it could be in a book, it could be in articles, but they have to have a way to know what they believe, to know what knowledge they have that is valuable, and find a way to add value to others.  I think that that’s just more and more the direction things are going. And it can be really hard because you are probably growing an empire, you are probably doing so many different things. And just know that there are people like Executive Presence, where our skill set is figuring out where your zone of genius is and figuring out what stories you have that are really good assets, and figuring out who you are and how you want to present online, and then helping you do that consistently across different channels.  And so if that’s of interest, I obviously love talking about this stuff, but I’m happy to talk with anyone who listens to the show. If you go to executivepresence.io and fill out our contact form, it gets to me. You can find me on LinkedIn, Justin Nassiri, or my email is justin@executivepresence.io. Any of those three work. But I just believe that this is going to be more and more valuable for leaders to develop that skill, and would love to help anyone listening do that.  Well, if you’re out there listening, you see that something is working because Justin propelled his company from a standing start to the Inc. 5000 in three or four years, and he is pushing the envelope on generative AI and reinvention every 10 weeks of his company. So if you’d like to be part of that and you have to take advantage of promoting yourself on LinkedIn with a cutting-edge approach, then reach out to Justin Nassiri on LinkedIn or executivepresence.io. And if you enjoyed this conversation, stay tuned because every week I bring a couple of successful entrepreneurs who are sharing their frameworks of how they’re being successful. So thanks for coming, Justin, and thanks for listening. Important Links: Justin's LinkedIn Justin's website Justin's Email: justin@executivepresence.io

This Week In Startups is made possible by: Vanta https://www.vanta.com/twist Agree https://agree.com YSecurity https://YSecurity.io/TWIST   Today's show: Frontier AI models can ace PhD-level exams, but it's still bad at tracking down the product you want in the style that suits you. Onton's Zach Hudson tell us that the problem is that models are a black box. His solution? A neurosymbolic model, Ontology 1, that learns about your taste and preferred aesthetic over time, then produces product searches tailored specifically to you, rather than just using keywords and relevant tags. How do neurosymbolic models work, and how does Onton understand your prompts and favorite design trends? And why aren't the frontier labs working on neurosymbolic models of their own? Zach joins Jason and Lon to discuss. PLUS, following a record-smashing SpaceX IPO, Ashi Dissanayake of Spacium makes the case that the real bottleneck in space isn't launching rockets off the ground any more. It's refueling in orbit. Guests Zach Hudson on X: ****https://x.com/nosduhz Onton: https://onton.com/ Spacium: https://spaceium.com/ Spacium on X: https://x.com/SpaceiumInc Relevant Links Poolside's journey to AGI: https://poolside.ai/vision/purpose Startup Archive: Sam Altman on the Paul Graham advice that saved OpenAI: https://www.startuparchive.org/p/sam-altman-on-the-paul-graham-advice-that-saved-open-ai-always-make-an-api Kelly Wearstler: https://www.kellywearstler.com/ Ennis House: https://franklloydwright.org/site/ennis-house/ Los Feliz Living: Ennis House profile: https://www.losfelizliving.com/los-feliz-historic-homes/ennis-house-los-feliz-hcm-149 Indiewire: Ennis-inspired "The Studio" offices: https://www.indiewire.com/features/craft/the-studio-production-design-interview-seth-rogen-1235114365/ Monocle Magazine: https://monocle.com/ Spaceium on Y Combinator: https://www.ycombinator.com/companies/spaceium-inc Orbit Fab's RAFTI: https://www.orbitfab.com/rafti/ James Webb Space Telescope: https://science.nasa.gov/mission/webb/ Houzz: https://www.houzz.com/ Timestamps: 0:00 Zach Hudson joins: What is "neurosymbolic search" 4:03 Ontology 1 isn't a black box 6:31 Who is using Onton? 9:36 Vanta - Get $1000 off your SOC 2 at https://www.vanta.com/twist 11:35 Could neurosymbolic models reach AGI? 13:19 Jason loves Wright's Ennis House 17:03 The shape of AI companies is changing 19:28 Agree.com - Stop chasing invoices and automate your entire contract-to-cash stack. Go to https://agree.com and tell them Jason sent you to get 50% off for life! 22:32 UGC as a data moat 26:03 The Dead Internet Theory 28:13 Ashi Dissanayake of Spacium joins 29:52 YSecurity - The on-demand security team for startups. Need enterprise-grade security without hiring a $400k CISO? YSecurity gives you 40+ expert engineers, matched to exactly what you need, by the hour, with your first six hours completely free. Go to https://YSecurity.io/TWIST 31:50 Storables vs. cryogenics: the zero-boil-off breakthrough 34:11 All kinds of propulsion requires refueling 36:09 Getting more value from LEO to GEO 38:28 Moving at rocket speed 44:54 Why demand is so acute 49:37 The investing climate for space, post-SpaceX Subscribe to the TWiST500 newsletter: https://ticker.thisweekinstartups.com Check out the TWIST500: https://www.twist500.com Subscribe to This Week in Startups on Apple: https://rb.gy/v19fcp   Follow Lon: X: https://x.com/lons   Follow Jason: X: https://twitter.com/Jason LinkedIn: https://www.linkedin.com/in/jasoncalacanis   Check out all our partner offers: https://partners.launch.co/   Great TWIST interviews: Will Guidara, Eoghan McCabe, Steve Huffman, Brian Chesky, Bob Moesta, Aaron Levie, Sophia Amoruso, Reid Hoffman, Frank Slootman, Billy McFarland   Check out Jason's suite of newsletters: https://substack.com/@calacanis   Follow TWiST: Twitter: https://twitter.com/TWiStartups YouTube: https://www.youtube.com/thisweekin Instagram: https://www.instagram.com/thisweekinstartups TikTok: https://www.tiktok.com/@thisweekinstartups Substack: https://twistartups.substack.com

Keach Hagey: Keach Hagey, author of The Optimist: Sam Altman, OpenAI, and the Race to Invent the Future, explores the rise of Sam Altman and the founding of OpenAI, which launched in 2015 as a nonprofit research lab aimed at developing artificial general intelligence safely. Altman partnered with Greg Brockman and lead scientist Ilya Sutskever, securing initial billion-dollar commitments from major players such as Elon Musk and Peter Thiel. The narrative follows Altman's trajectory from a brilliant student at John Burroughs School to a Stanford dropout who founded the startup Loopt. Though Loopt was considered a relative failure, Altman's charismatic storytelling and investment prowess eventually led him to succeed Paul Graham as president of Y Combinator. As OpenAI's needs for computational power grew, the organization transitioned into a complex for-profit structure, leading to a power struggle that saw Musk depart. The account highlights a pivotal 2023 crisis in which the board fired Altman over concerns regarding his transparency, only for him to be reinstated after a massive staff revolt. Throughout, the book balances Altman's unwavering optimism for the future against stark warnings from AI godfathers about the potential existential risks of unaligned artificial intelligence. (1)

Enterprise account executives are now signing for $200,000 salaries, $200,000 variable comp plans, and up to half a million dollars in equity at sign-on. Meanwhile, offer acceptance at the top AI companies sits at 45% and half of the 38,000 startup sales reps in San Francisco and New York have taken a new job in the last two years. In this episode, Sam Jacobs, AJ Bruno, and Asad Zaman go hosts-only on why the sales talent market broke, what QuotaPath's compensation data across more than 1,400 companies says about quota attainment splitting into a barbell, and why Paul Graham calling go-to-market bogus says more about Silicon Valley's blind spot than it does about sales. Plus, where young sellers still get their start, whether a CRO can be great without ever carrying a bag, why go-to-market cannot rescue a company that has lost product-market fit, and what HubSpot's 20% stock drop signals for every SaaS business trying to make the AI transition. Key Takeaways: - The AI buildout has drained the two talent pools it depends on, and compensation is repricing in real time. As Asad Zaman, CEO of STA, described the enterprise AE market: "Enterprise account executives now get paid $200,000 salaries, $200,000 variable comp plans, up to half a million dollars in equity at sign-on. That used to be $150,000 to $175,000 at the top end, just like a year and a bit ago." With 38,000 startup software sellers in San Francisco and New York and a quarter of them changing jobs in the last twelve months, the constraint on AI companies hitting escape velocity is no longer capital, it is people. - The jobs data cuts against the automation narrative. Sam Jacobs, CEO of Pavilion, pointed to the Philippines outsourcing sector, 8% of that country's GDP, where employment in IT and business outsourcing is up 20% to 1.9 million workers and industry revenue is up 30% to $42 billion since the launch of ChatGPT: "I just think, you know, it's Jevons' paradox. It is what happens with every new piece of technology. Everybody thinks the technology is going to wipe everybody out, and instead the economy adapts." Companies are leaner per dollar of revenue and still cannot hire fast enough. - Quota attainment is no longer distributed the way comp plans assume. QuotaPath's first-half numbers put 58% of AEs on track against annual quota versus 55% a year ago, but as AJ Bruno, CEO of QuotaPath, explained, the shape underneath moved: "the standard deviation is way higher this year. Meaning that the winners are going to be winning more … So it's more like a barbell than it is a bell curve." At some AI companies two reps out of ten are closing half the entire number, which breaks the capacity math most sales leaders use to set targets. - For individual sellers, the window matters more than the title. Asad Zaman's advice to reps weighing a step up into leadership: "this is the moment where you can make millions of dollars right now … So don't make silly choices … Become a leader later. Go make money right now." He now benchmarks offers on whether 300% of plan clears a million dollars, and warns that private equity portfolio companies that never gave equity to individual contributors are losing these candidates outright. Connect with the Hosts: Host: Sam Jacobs, CEO at Pavilion - https://www.linkedin.com/in/samfjacobs/ Host: AJ Bruno, CEO at QuotaPath - https://www.linkedin.com/in/ajbruno3/ Host: Asad Zaman, CEO at STA - https://www.linkedin.com/in/azaman1/ Topline is more than a YouTube Channel: Subscribe to Topline Newsletter: https://toplinemedia.substack.com/ Tune into Topline Podcast, the #1 podcast for founders, operators, and investors in B2B tech: https://www.joinpavilion.com/topline-podcast Join the free Topline Slack channel to connect with 600+ revenue leaders to keep the conversation going beyond the podcast: https://www.joinpavilion.com/topline-slack Chapters:  00:00 Three Hosts, Three Topics 02:19 The Talent Market Has Gone Crazy 05:42 $200K Base, $200K Variable 09:00 AI Was Supposed To Kill Sales 15:16 Go Make A Million Right Now 16:53 Where Young Sellers Get Their Start 23:20 Build A University Inside The Company 31:58 Paul Graham Versus Go-To-Market 36:22 Why YC Misses Enterprise Software 40:55 Can A Non-Seller Be A Great CRO? 44:42 GTM Cannot Fix Product-Market Fit 49:30 One To Ten To A Hundred 53:38 Quota Attainment Is A Barbell 1:02:41 The HubSpot Nightmare 1:06:05 Falling Back In Love With The Business

$0 -> $100M ARR. How Gamma Scaled Quickly without a Sales Team 100 million in ARR. A team of 50. Zero sales reps. Grant Lee, Co-founder and CEO of Gamma, shares the exact playbook behind one of the most capital-efficient growth stories in SaaS - from pitching investors out of a London kitchenette to going viral with a single tweet that got Paul Graham throwing shade. In this session, Grant breaks down four lessons: 1. Product-market fit isn't a checkbox. After winning Product of the Day on Product Hunt and watching signups plateau, Gamma went back to the drawing board. They gave themselves three months to make the first 30 seconds of the product feel magical - and word of mouth did the rest (5K signups/day, then 10K, then 50K, zero marketing spend). 2. Creator marketing only works if you've done it yourself. Grant went through "Cringe Valley" to understand what creators actually need - then used that to manually onboard every creator partner and build something that felt authentic, not transactional. 3. Community-led growth is literal. At 50M users, Gamma flew power users to SF, visited customers in Seoul, London, and São Paulo, and created a Gambassador Slack where early feedback shapes the product roadmap. Your users are not a faceless entity. 4. Dogfooding the future builds conviction. How Gamma killed their virtual office idea after six months and went all-in on presentations - and why testing your own product is the fastest path to knowing what to build next. If you're building a product-led company and wondering whether to invest in marketing or go back to the product - watch this first.

In recent months, the open vs closed, and US vs China discussions on model ownership and sovereign/local AI have heated up to a fever pitch. So it is very very good news that Poolside AI are finally emerging with new models, like Laguna S 2.1, that are beating Thinking Machines' recent release nearly 10 times their size.Poolside's recent tech report got a lot of praise due to their level of detail, and Vibhu first covered Laguna's recent technical report on our paper club:From spending $12 million building language models for code before the world cared to creating a Model Factory that can take a model from pre-training to release in eight weeks, Eiso Kant has spent more than a decade betting that code is the path to AGI. In this episode, the Poolside co-founder joins swyx and Vibhu to explain why ChatGPT felt like vindication, why Poolside embraced open weights and open research, and why he would rather live in a world with 100 foundation model companies than five even if Poolside were one of the five.We go deep on Poolside's Model Factory: the engineering systems behind 10,000–20,000 experiments per month, streaming data directly into training, reproducible experimentation, low-precision compute, and agents that increasingly write code, launch jobs, evaluate results, and modify the pipelines used to train future models. Eiso also unpacks their recent launch Laguna S, why persistence, verification, and backtracking may matter more than raw intelligence, how much capability remains inside smaller models, why reinforcement learning will move earlier into pre-training, and why next-token prediction is still extracting too little from the web.We also discuss model-harness co-design, Poolside's path from coding agents to AGI, why Eiso thinks MCP and traditional tool calls are “stupid,” the real economics behind frontier-model training, Poolside's $500 million raise, open-source AI, regulation, NVIDIA and TSMC's influence, engineering productivity in the agent era, high-agency teams, and hiring at Poolside.We discuss:* How Andrej Karpathy's RNN work inspired Eiso to start building language models for code in 2015* Why Eiso spent four years and $12 million pursuing an idea before the market cared* Why ChatGPT felt like vindication and brought Poolside back to open source* Why Eiso would prefer 100 foundation model companies over an oligopoly of five* The difference between releasing open weights and publishing genuinely open research* Why Poolside deliberately built a global research organization outside the Bay Area talent war* Why model building is ultimately 90% engineering* The Model Factory: Poolside's end-to-end system for rapidly training and improving models* How fewer than 70 researchers run roughly 10,000–20,000 experiments each month* How Poolside moved from six-month model cycles to five- and eight-week launches* Why streaming data directly into training unlocked faster experimentation* How immutable data, versioned code, and reproducibility enable rigorous model research* Why Eiso wants capable researchers to leave their labs and become Poolside's competitors* Why 95% of model building can be reduced to better data or compute efficiency* Laguna S and why persistence, verification, and backtracking can outperform raw intelligence* Why smaller models may handle far more knowledge work than previously expected* Why reinforcement learning will move earlier into pre-training* Why next-token prediction is still failing to extract enough knowledge from the web* Why distillation and environments have become the AI industry's favorite “drugs”* Why mid-training is really an early form of curriculum design* Low-precision training, networking bottlenecks, and the next gains in compute efficiency* Laguna S: 118 billion total parameters, 8 billion active, and eight weeks from training to launch* Why model builders can often evaluate a new checkpoint within its first 30 minutes* Model versus harness: where agent capabilities actually come from* Why Poolside sees coding and long-horizon software tasks as a path to AGI* Why Eiso thinks MCP and traditional tool calls are “stupid”* Why future agents will write scripts instead of choosing from dozens of predefined tools* The case for minimal harnesses, containers, and model freedom* Why Poolside is prioritizing vision but does not expect to work on audio soon* Why language may be the most compute-efficient modality for encoding knowledge and reasoning* The real cost of model development and why the final training run is anticlimactic* The story behind the Poolside name and why it represents refusing to lower ambitions* How Poolside raised $500 million while investors still questioned whether AGI was real* Why intelligence could become the world's most demanded and commoditized resource* When open models may become too capable to release without restrictions* Why unilateral AI safety does not work in a globally competitive environment* How regulation could accidentally lock in an oligopoly of two or three AI companies* NVIDIA, TSMC, and the hardware systems underpinning foundation-model progress* Why reinforcement-learning wall-clock time is one of Poolside's biggest bottlenecks* Why Poolside trains models from scratch instead of simply distilling larger models* How AI changes the way companies should measure engineering productivity* Why agency may become the most important quality for employees in the AI era* How leaders align high-agency people through shared goals and clear constraints* Hiring across research, post-training, pre-training, architecture, evals, and engineering at PoolsideEiso KantLinkedIn: https://www.linkedin.com/in/eisokantX: https://x.com/eisokantPoolside: https://poolside.aiTimestamps00:00:00 Introduction00:00:54 Karpathy, RNNs, and Building Code Models Before Transformers00:02:26 The $12M Failure and ChatGPT Vindication00:03:39 Open Source and the Case for 100 Foundation Model Companies00:09:22 Open Weights, Open Research, and Poolside's Global Team00:16:04 The Model Factory: Why Model Building Is 90% Engineering00:20:19 Agents, Automated Experiments, and Early Signs of RSI00:24:04 Streaming Data, Reproducibility, and Scientific Rigor00:30:35 Creating More Foundation Model Companies00:36:07 Laguna S: Persistence vs. Raw Intelligence00:43:01 Reinventing Pre-Training, RL, and Curriculum Design00:52:33 Low-Precision Training and Squeezing More From Smaller Models00:58:37 Model Harnesses, Coding Agents, and the Path to AGI01:09:26 Why MCP and Traditional Tool Calls Are “Stupid”01:13:04 Vision, Multimodality, and Why Language Still Matters01:18:15 Scaling Models and the Real Economics of Training01:20:40 Why Poolside Is Called Poolside and Raising $500M01:27:37 Open Models, AI Safety, and the Risk of an Oligopoly01:33:53 NVIDIA, TSMC, and the Reinforcement-Learning Bottleneck01:41:52 Smaller Models, Distillation, Engineering Productivity, and HiringTranscriptIntroduction: Eiso Kant, Poolside, and Open ModelsSwyx [00:00:00]: All right, we're here in the studio with Eiso Kant from Poolside, together with Vibhu. Welcome.Eiso Kant [00:00:08]: Thanks. Thanks for having me, guys. Good to be here.Swyx [00:00:10]: Yeah, fresh on the plane. You texted me, you were like, “Hey, I'm on my way to SF.” I was like, “You're on a plane right now, right?” Like, hey.Eiso Kant [00:00:16]: I know. After I texted you, I realized that probably coming in with major jet lag was gonna offer some fun experiences today, but let's do it.Swyx [00:00:23]: I mean, I think the thing I would tell guests is that they don't have to prepare that much because if you're truly working on this every single day, then even, like, what you hazily remember is going to be new for a lot of the audience that don't live in your world every day, right? so 10 years ago, you did a talk at Google Slush, talking about the democratization of AI. and, now here you are, like, open sourcing an incredible new model that we're gonna talk about. But I guess, like, what got you into democratization of AI? Like, it's not obvious from your LinkedIn or something.From Karpathy's RNN Post to SourcedEiso Kant [00:00:57]: No, it's not at all. I don't think it's obvious how I got in this space. I owe getting into this space to Andrej Karpathy.Eiso Kant [00:01:05]: In 2015, he wrote an article called “The Unreasonable Effectiveness of Recurrent Neural Nets.”Swyx [00:01:10]: Neural Nets, yep.Eiso Kant [00:01:11]: And that article, I read it, and I pivoted my startup at the time overnight to working on RNNs, and later LSTMs and Transformer models to be able to write code. If you go to this article and you scroll down, you can start seeing, like, this was the precursor to what ended up becoming language models. So, at least when he was character-level language models that were starting to predict letters, he has an example out here. There's a little Paul Graham generator, and you can read it, and the text makes sense, but it doesn't. and there's a little-- There's an example of code a little bit further down. Yeah, so Shakespeare.Swyx [00:01:47]: Shakespeare.Swyx [00:01:49]: CoolEiso Kant [00:01:49]: And for some reason, I read this, and I went down the rabbit hole of learning everything I could about RNNs and LSTMs, right? This is Transformer paper. And I had built a completely unreasonable belief, that neural nets should be able to generalize to anything and everything, and that language should be able to generalize, to a lot of things that are intelligent and the ability to write code. And so I started building Sourced, which was a fully open source company trying to build, what we used to call machine learning on code, language models on code. And we spent about four or five years on this, till the end of 2019. And that sounds really cool today, but back then, no one cared.Eiso Kant [00:02:29]: Right? Like, no one cared. We were in the dark. Like, we did things along the way. We tried applying convolutional neural nets to, like, the structure of code. We were. when attention came out, we were applying it to LSTMs, and then the Transformer paper came out. And it - it wasn't obvious, and what we missed throughout that entire journey, that we were on the right track, but we should have just kept scaling up. And today, to all of us, the scaling laws and scaling up seems like the most obvious thing. But having spent four or five years of my life on working on language models on code, it wasn't obvious. So I have a lot of respect to folks at Google and OpenAI and others who took that confidence and kept going. we failed ultimately at the time, and it was, like, biggest failure of my career, right? You blew $12 million of investors' money, which was a lot back then.Swyx [00:03:18]: Yep.Eiso Kant [00:03:19]: You spent, still a lot, but, And you spent years with, like, a group of 40 people just obsessing over this problem. And life took a different turn, And it was, and family became a focus, and I kept my heads down and really, didn't really look at language models for the following two years. big mistake considering Following years are gonna be really interesting. And then ChatGPT came out And it was like a vindication. It's like people started texting me. I found, like, my old, work decks and these old talks. And throughout that whole journey, we,ChatGPT, Vindication, and Returning to Open SourceEiso Kant [00:03:56]: We really had a strong point of view at the time that, like, as you're building more capable intelligence, it should be open and open source.Eiso Kant [00:04:04]: When we started Poolside, that wasn't the case at all, and I wanna be very open about it. When we started Poolside, we were like, there was a premise of two things. One is this technology is not gonna stop compounding in capabilities. I think to most people obvious today, but three-plus years ago when we started, most people were still arguing if these were stochastic parrots or not.Eiso Kant [00:04:23]: And the second was that reinforcement learning was gonna be the biggest driver for LLM capabilities. Today, very obvious. Three years ago, was not an opinion held or direction held at either OpenAI or Google or Anthropic or others. And so people looked down on us a little bit. They were like, “ is this really gonna work?” And so we just started working the problem, and we never really thought about open source again. We just kept our heads down and we built our, like, knowledge, understanding from scratch, right? We didn't roll out of an existing lab. So we picked up the papers and started writing code and figuring things out.Eiso Kant [00:04:59]: And it wasn't until the beginning of this year that me and my founder, Jason, picked up the open source conversation again.Eiso Kant [00:05:07]: And if you go back to some of the early things on our website, it was very straightforward. It was we wanna get to AGI, we wanna support a world of abundance, and we wanna be the first company that gets there.Eiso Kant [00:05:20]: But we started talking at the beginning of this year because it became obvious that the world was going in a direction that was starting to like, pick at us a little bit. Like, it didn't, this didn't happen overnight. It was, like, a little bit we were seeing this and we're like, “Okay, The world's going down a path.” And Throughout this journey, there was something that I used as a, as an analogy or thing. So I said well, if I go back to back in those days, 2015 or 2016, we're working on this, and I picked up a fi book off the shelf, and I was reading the book about 2035. AGI is achieved, and the story would be over the following, decades. And it would have that first chapter where everyone's trying to figure things out. You'd get the chapter of ChatGPT coming out And then you would get to the chapter where the world was at a fork in the road, and the one that it picked was one where three or four or a handful of companies were going to create all of intelligence moving forward.Eiso Kant [00:06:21]: And when I thought about that story, it felt like a dystopian fi book, not a utopian fi book. And the reality is, I'm a utopian fi guy. Like, and so We took a step back and said, “Hey, can we play a role here?” Now it was easy for us to do so because we were not at the frontier.Eiso Kant [00:06:41]: If we were at the frontier, I don't think we could have changed our mind. and I don't mean this like it's when the moment there's too much capital involved, too much expectations, you've built up things, right? We're a small team, just improving and improving. And so we knew that we could make that decision now, but it would be a lot harder to make as we got closer and closer to the frontier and caught up to others. And did a lot of soul-searching and a lot of conversations, and said, “No, this makes sense,” Even if there's big unanswered questions, like how the hell do you build a business model with foundation models about open source? Big open-ended question that we do not fully have the answer to yet, right? At what point do you no longer wanna release open source models because misuse of models has, real potential risks associated with it? how is the government gonna respond to open source? but I think it all just came down to one thing, and I'll stop the monologue, is the fact that I rather live in a world that has 100 foundation model companies than a world that has five, even if I was one of the five. And the smallest and most meaningful contribution we can make for 100 to exist is to open up our research and open up, like, our weights right now and figure out along the way how we can, like, do more.Neo-Labs, Model Choice, and the Token EconomySwyx [00:08:01]: Yeah. I think if anything, over the past three years, that has become a bit more true. you are one of a cohort of Neo labsEiso Kant [00:08:10]: YeahSwyx [00:08:10]: That people are now calling that. And, we're, we're doing this on the day that Thinky launched their, new model and you are outperforming them on their, on some benchmarks that they released, right? Like, they just don't have it yet. so it goes to show that I think, like, this is one of those things where, like, there is room for multiple players, and you are seeing a little bit more of the future. Maybe more like 20, not 100, but, like, you are one of the 20.Eiso Kant [00:08:36]: I really hope so, right? I think we I'm, I'm excited about their release, and I'm excited about everyone releasing because, like, ultimately, like, choice competition is both gonna drive progress in the right direction. But the fact that like, we create models and while we all, drink out of the same well of data effectively, we do introduce very different behaviors and biases in our models. Some are intended biases, some are completely unintended biases.Swyx [00:09:03]: Yeah.Eiso Kant [00:09:03]: And if we shape up in an ecosystem in the world where open models are gonna be a part of the token economy, like, I don't think there's any question about it anymore Then we want to be able to live in a world where companies, countries, people can choose and say, “Hey, I am most aligned and I trust most this provider for these things.”Swyx [00:09:25]: Yeah.Vibhu [00:09:26]: I think more than just one of the 20 Neo labs, up until recently, most of open source innovation was coming from the Chinese labs, right? So there's the DeepSeek of the West. Is it today? Okay, maybe it's thinking machines reflection, but there aren't many, right? So, one of the things you guys started in France, Europe, but very much now you're taking that American standpoint and more than just that, the point is the Chinese models that we see, they're not super open research. the work you put out is, I think, some of the best. So every few months you get not only frontier models, but also here's a breakdown blog, paper, technical report of here's everything for state of the art to build, frontier intelligence and you're filling that gap too, right? So not just only open weight, not just Western, but also pretty open research.Open Weights vs. Open ResearchEiso Kant [00:10:20]: No, I appreciate it. Look, I think it's, I think it's the most meaningful contribution, right? Weights are a binary. Let's call them what they are. Yes, we can modify them, we can change them, but, like, giving someone the weights does not allow them ultimately to recreate what you're doing, right? And so now there's challenges around releasing data sets, challenges around like releasing certain things, but being able to share your research, like, right, how do we do it? What are the lessons we learned that we spent, tens of thousands of experiments of compute on? I think very much so. One correction though, Vibhu, and I say this because it's been haunting us for quite a few years. We from day zero were an American company.Swyx [00:10:55]: Yeah. They movedPoolside's Global Team and American Company StorySwyx [00:10:56]: To France.Eiso Kant [00:10:56]: So the story once and for all is very. We start as an American company. We have always been an American company, and early on we made a very conscious decision. We said, “We're not gonna hire any researchers in the Bay Area. We're gonna look for talent everywhere else in the world.” and that is everything from Middle Americas, Seattle to, Serbia, and to Taiwan and Singapore and other places. And it was because we took a view that this was gonna become a talent war for this, and I think it has over the years now. Three years ago, that wasn't fully obvious yet. I think today it very much is. And we also realized that, like, some of the world's most capable people with, like, the most interesting, innovative ideas were not just gonna be here. And so it led us to create like a fully remote company. and we ended up opening an office in Paris and London and different places and we have a lot of the team in the US and a lot of team outside. But we always took this view of like, we're an American company, but if we want the best of the best to work with us, we need to take a global view. Now we do also have people here in Silicon Valley, like the company's grown and others, but I think one of the things that, it slowed us down at the beginning, but it has sped us up now, and it's why you're seeing like the progress, I think, on our models and the cadence at which we release, is because we didn't roll out of an existing lab. Right? we didn't, we didn't have a lot of the information that's freely flowing around here at the time. We just took this point of view as like, “Okay, well, let's just work the problem. Let's just go and, like, read the few papers that are out there, and let's just figure this stuff out.” And we made some hilarious mistakes in model training because of that over the yearsEiso Kant [00:12:35]: Like especially in the first 12 months. there's a few that I think still haunt me and scare me. We can talk about them later. but it created a, like, a resiliency and persistency in the team, right? with extremely few people have left us over the years, that, like, told us, “Okay, we can do this.” When we first wrote our first training code base completely from scratch, it wasn't a fork of any open source. It was just like, “Okay, let's build it from scratch.” I remember we had this one moment where we spent three weeks working out an optimizer bug. Like, it was like training just couldn't get stable. We, like, obsessed over it, and we thought, like, maybe we were wrong. Maybe we should have just forked this repo, or we should have. But then when we solved it, I still remember at the time we were like five people in the company. when we solved it, we were like, “Oh, we can do things,” like if we're just willing to work hard. and I think that culture with a very strong engineering bias has helped us, like, get to where we were. And so there's this notion of open source and talent and these things. I think we, We just took different decisions from a different starting point. and I think we are lucky. I do want to definitely call it lucky. And there was a lot of hard work at the team that now, like, that's starting to show up in results.Swyx [00:13:52]: Just ‘cause we probably won't revisit this again, but, and this is a fun recruiting challenge if someone knows the answer. What was the bug? And then we won't tell the solution, but we'An Optimizer Bug and the Value of Building From ScratchEiso Kant [00:14:01]: So the - This - You're gonna test my memory here,Swyx [00:14:04]: Oh, okayEiso Kant [00:14:04]: So but I thinkSwyx [00:14:05]: DirectlyEiso Kant [00:14:05]: I think I can recall. So if you, so if you look at, So if you take like Adam as an optimizer, you have epsilonSwyx [00:14:12]: YeahEiso Kant [00:14:13]: Which is, right, like in the denominatorSwyx [00:14:14]: Momentum and weights. YeahEiso Kant [00:14:15]: Is exactly, in the denominator. And at the time, if I recall, you looked at like the early Llama papers and things like that. People were juicing epsilon, like, quite a bit. Like, they were, like, adding, I don't know if it was E minus four or whatever, like a high value for epsilon.Eiso Kant [00:14:31]: And if you think about this during training, it's like a bit weird and counterintuitive that we're adding noise to our optimizer by just adding effectively, like, a random number in the denominator, right? Like behind the decimal point. And I don't recall the exact bug, but it had - What I remember is once we solved it, we no longer had to juice epsilon as much as, like, was happening in the Llama paper and other places. and it was like one of those fundamental moments where we had trusted this paper that was out there, and we're like, “Oh, no, it has to be this way. It has to have this high value of epsilon.” But it made no sense to us intuitively. Like, why do you have to have this so high? Like, if you're just trying to avoid division by zero, why can't the value be extremely small? and that was like one of those moments where you realize like, okay, finding things out from scratch yourself builds a better intuition. Because the one thing you learn very quickly with model building is that your intuitions that you start with are gonna get beaten up so hard.Eiso Kant [00:15:33]: Right? Like - It's such an experimental science, that the things that seem obvious, you very quickly get to learn, like, you were wrong, and hopefully you figure out why, and sometimes you don't even.Swyx [00:15:45]: Yeah. yeah, so, one of the reasons that you, when you released your new models, Vibhu got really excited. I mean, everyone got really excited. But Vibhu led our paper club on it, and you guys sawEiso Kant [00:15:58]: YeahSwyx [00:15:58]: Obviously. maybe talk through some lessons learned in that, whatever you can disclose. we can focus on the model factory stuff, whatever you think is a good starting point.Model Building as EngineeringEiso Kant [00:16:08]: So I would say that our view from very early on in the company was that model building is ultimately 90% engineering.Eiso Kant [00:16:18]: And I think we all know it in the industry because if you look at where's every researcher spending their time, they're spending their time writing code, right? Looking at data and writing code. And so we said, okay, The state at the moment, like three years ago, was bash scripts and Slurm and spaghetti code bases for training and, like, data pipelines that were patched together. And we looked at this and said, “Well, ultimately, model building is a process.” You're going from raw data, right? Like training raw material, the web, et cetera. you're doing a whole bunch of filtering, cleaning up, transformations, analyzing. These days, that's, far more complex than it was three years ago. then you're training a model, which is effectively a large distributed systems problem, right? Across hardware that has still-- It's become a lot more reliable. It was extremely flaky back then. and now with every new generation, we get our new sets of challenges. And then you go into the next stages, right? There was no training back then, but, like, you got, your post-training and then your reinforcement learning. And so we looked at this and we said, “Well, this looks like an industrialized process. This looks like an end process, that every single part of it has its machinery,” right? If it's your big data pipelines, if it's your crawling ingestion of the web, if it's your, large-scale distributed training, and then you've got your reliability. And we said, “Well, why don't we take some of the world's smartest distributed systems engineers that we knew and make them part of the process of research from day zero?” Not retrofitting it later on, but, like, really from the beginning. And that became our model factory. And so our model factory started with a handful of components. Today, it's thousands of components, and I try to equate it to, if you think about, like, someone who was at the very early days of Foxconn, if they had been there for the following, decade, they would be able to rebuild Foxconn because they saw every decision that led to building that system and all the complexity. If you and I walk into Foxconn today, no chance.The Model Factory and Experiment VelocityEiso Kant [00:18:18]: Right? Because we don't have the lineage and history of decisions that led to that. And so we built early on from the beginning- with a team that really understood that, well, the metric that we are optimizing for is the speed of an idea from a researcher to an experimental result that we can trust to then being part of the next model training.Eiso Kant [00:18:42]: And in the. And because it's such an experimental science, ultimately, in the beginning when it wasn't that complex, you could patch your way around it, right? But now, at any foundation model company, you are running. I mean, we're a small team, right? We're less than 70 researchers, another 35 engineers. and we are running, I haven't checked the latest count, but far more than 10,000, maybe 10 to 20,000 experiments a month that we cut. And so if you look at that scale of every model run that is, like it's ultimately it's, it's you need to be able to trust it as an infra problem. And so what we have now done over the years is gotten really good at that, and just by working it and improving it and obsessing over those end decisions. So now what that means is that you looked up Laguna XS 2 that we launched. It was five weeks from the beginning of training to launch. The model that we're gonna talk about today was eight weeks from start of training, to launch. We started the next model literally yesterday because we now finished the post-training required for the model we're launching, next week or by the time this comes out today. and we move that compute to the much larger Laguna M model that we're now training. And so the model should be an artifact of someone's process. It shouldn't be really a thing in itself. Like, and we treat this like the way you would look at like a SpaceX factory where, yes, the first rocket, really hard to build, but the much harder challenge was building the factory. And now they're rolling off, and no one is really thinking about the next launch anymore. So it's just another launch, it's another launch, another rocket comes off. And that's what we're trying to do with model building.Eiso Kant [00:20:22]: And what has been, which was not planned from day zero, it was in the back of our mind like this will happen one day, is that when you build a really good end model factory with really good APIs and really good engineering systems, Well, what is it perfect for? It's perfect for agents.Agents Inside the Model FactoryEiso Kant [00:20:40]: Because agents are now starting to take over more and more work in our model factory.Vibhu [00:20:43]: Yeah.Eiso Kant [00:20:44]: So I look at the screens when I walk, like when we're, we come together, in our monthly, we do monthly onsites, and I walk behind people's screens and I stop by and I talk to our researchers. And the default is all of these different agents running on their screen that are writing the code. They're launching the jobs. They're evaluating the results that are coming back from the model runs. They are, making the changes. And we're still in the driver's seat. We're still coming up with the ideas. We're still helping with the debugging. But more and more, and this is right now very profound on the data side of our pipelines in both pre and post and the synthetic data pipelines, it's starting to become more on the architecture side as well. You're starting to see these twinklings of what RSI is gonna look like.Eiso Kant [00:21:27]: And that's. So when we talk about, like to your question about our models, every talk about the model factory, And my coolest example of these things is always that when we kick off a new run, doesn't matter if it's a training like big run or if it's now a post, like one of 10 post-training versions we do for like release or many experiments, is that at any given moment, the changes that somebody made that they had experimental results from the day before make it into that run.Eiso Kant [00:21:57]: So there's not like a cutoff 90 days before. Like no, it's like literally from that moment because we can now trust the machine enough. And then you also have to invest in the reliability. So one of my favorite metrics about like Laguna S is that there was no call events, Right? Like completely zero. And we haven't had a meaningful call event, like something to wake up for, as far as I recall this entire year. now there is one asterisk to that. In usually the first six hours of launching a new model run, something breaks because you set a config wrong, you made a small mistake, et cetera. So that's usually there's a little bit of intervention, but that's always within like call periods, right? Not on call. And I think that's starting to now compound. So the model we're releasing now, I love it. It's amazing, but we're already onto the next one. and I think that's the way it should be.Laguna, Five-Week Builds, and Zero On-Call EventsVibhu [00:22:50]: Hey, I also just wanna point out, so for context, this was like a month ago. we found it in the tech report, so we just came in with, “Okay, new model's dropped. Haven't heard about it.” We wereEiso Kant [00:23:02]: Yeah, we're very used to doing this every few months.Vibhu [00:23:03]: We're, we're very much like, “ okay, look, it's like, on par with Kimi, DeepSeek, whatnot, the small ones, Gemma level. Oh, it's a very cool paper on what goes into building.” And then we hit this page, right? Like literally page two of tech report is, “This process allowed us to build the small model from scratch to delivery within five weeks applying the lessons”. And then I'm like, oh, this paper is not about here's a tech report of benchmarks and here's how many tokens it was trained on. Like for people that wanna dive more from what we're not gonna discuss on the podcast, it's all laid out here, right? FromEiso Kant [00:23:38]: YeahVibhu [00:23:39]: Custom software that agents can use to interface with training code, training data.Eiso Kant [00:23:45]: Yeah. Well, link the paper correctly, so yeah.Vibhu [00:23:47]: Yeah. All that stuff. read the paper here, but,Technical Report Principles and Streaming Training DataEiso Kant [00:23:50]: But I would like to. I love principles, and I think that is a good starting off point for maybe telling some stories. Maybe we can go one by one past the principles. I'll just call out that Dagster just got bought by a Prefect.Vibhu [00:24:01]: Yeah.Eiso Kant [00:24:01]: Isn't it fun? But yes, I'm very familiar with Dagster. just anything where like they trigger some story.Vibhu [00:24:07]: So, well, I would say, well, experiments code's obvious, but I think one of my favorite things is, I don't know where it is in here, but early on, and I still think this is the case a lot of foundation model companies, people prepare their training data sets, they get packaged up, then they get copied over to a training cluster distributed across all of the nodes, and then training starts.Vibhu [00:24:30]: And we looked at this like three years ago and we were like That makes no senseEiso Kant [00:24:36]: You lose so much time because the moment you have to rematerialize the data set, you have to make a change, you have to fix something, et cetera, you've got all this time of like repackaging it, right? Toca- tokenizing it, repacking it, moving it over to a cluster, then distributing it across the nodes. The bigger your clusters are, you start using fancy like torrent-like algorithms to like distribute your data. So why aren't we streaming data into training? Right? Something that's very common and like just basicVibhu [00:25:00]: Like just in timeEiso Kant [00:25:01]: Just in time, like good computer science like principle. And that was one of the first things that I think unlocked - the model factory. Because the moment you start thinking about, well, a training job, it doesn't matter if it's a big hero run or a small like, post-training experiment, consumes a certain number of tokens per second, right? And it's not a lot, right? From a like a data, moving data perspective. So we said, well, we have our training cluster, and then we've got like our AWS kinda setup where we can build these amazing big data pipelines. We can set things up. We use Spark underneath the hood, like all these things.Vibhu [00:25:36]: But when you say AWS, it's not actual AWS, it's your internal AWS.Eiso Kant [00:25:39]: It's our internal-- No, it's our internal like just running like our infrastructureVibhu [00:25:42]: Site web servicesEiso Kant [00:25:43]: Exactly. Our stuff running on like an AWS account or on like any hardware, right?Vibhu [00:25:47]: Yeah.Eiso Kant [00:25:48]: And so once we made that shift into I can stream data into training, all of a sudden you realize a lot of things unlock. Because now you don't have to wait for the whole data set to materialize.Immutable Data, Experiments as Code, and Scientific RigorEiso Kant [00:26:00]: You now all of a sudden when you're running data experiments about mixing data, it's a config. Because you've got these data sources that are coming in, and you just - we have this service called Blender that's in the report, where we then say, “Okay, for this run, I want 20% of this source, 10% of this source. I want this much, so many epochs of repetition. I want this to be, shuffled in a certain way,” and your training job can start while the rest of the data is even still materializing. also what it does is because all of this underneath-- So for us, we treated the data layer underneath as like an immutable data layer, and that was really important. Like experiments as code, immutable data layer means that you can always go back and understand literally down to the single token at which cursor it went in on which version of the code.Vibhu [00:26:47]: Yeah.Eiso Kant [00:26:48]: And it took us a I have to admit, like the first year of Poolside, we understood that engineering had to get great, But we didn't understand yet, that this is ultimately in support of like a good rigorous scientific progress. We were quite a - We were a very small number of people, so a lot of it was YOLO ideas and YOLO runs.Vibhu [00:27:08]: Yeah.Eiso Kant [00:27:09]: And we built great infra for the YOLO runs. But once we realized that we treated data as immutable and code as always versioned, and you could always track and trace every experiment end to end perfectly, you could repeat everything perfectly, right? You have perfect reproducibility. I can still reproduce runs from two years ago if I wanted to, right? It enables the scientific progress, like the scientific process, and I think that took us probably about a year and a half into the company to figure out. We also had some great hires, like our head of applied research, Nikolai, who joined us from Yandex, who'd been working on language models since like the early 2020s, I think brought that into the company of like, “Hey, we wanna have even more rigor.” And then once we kinda had the combination of like increasingly more capable platform that allowed people to do more, but had this immutability, we were able to start “Okay, every experiment is truly an ablation. We truly need to understand it.” And I think we became much more scientifically rigorous in the last couple of years, and the infra underneath enabled it. and then there's just fun stuff like, andVibhu [00:28:16]: Yeah, a lot of it's fun, like even just the, one, you share all the ablations, two, picking the data sets, right? There's like a random small paragraph in here where it's just like, “Oh yeah, training data, we have some, we have an auto mixer.” it trains eight small models, scales them up, picks the training data set. We don't even need to look at it. I'm like, “Wow, a lot of engineering rigor there.” And there's just, there's just a lot in here.Publishing Research and Giving BackEiso Kant [00:28:40]: Yeah, and it'- and look, and we wanna put out more. Like we, We treat writing papers as something that we haven't earned the right for yet for a long time. So you earn the right to spend time, publishing research once you're at the frontier, because until then, you're catching up, and every minute and hour in this industry matters. Like I obsess over it, not just the wall clock time from idea to result, but just general like time every day that we, waste is one that doesn't allow us to catch up. But in this case, we said, “Okay, we're gonna give ourselves.” I think we gave the team like three or four days while still doing their work, like give everything in there. And to your point earlier, if your stuff, it's easy to like put it out. And so there's so many more things that we wanna talk about over time, and we will definitely start doing. And as we earn more of the right, but also now have like added to our mission that we want more foundation model companies to exist, you'll see us like be way more proactive, and just trying to keep dropping some of those like things that we've learned along the way that can help others like speed up.Vibhu [00:29:40]: Which is the other cool side of this, right? It's, it's not like, back to your point, it's not just here's the benchmarks of our training. If you want to replicate, here's experiments of optimizers, data sets, post-training. you lay out a lot of it here alongside here's your system for how to do it? So it's, it's really like promotingEiso Kant [00:29:59]: No, thank youVibhu [00:29:59]: Other people can do the same.Eiso Kant [00:30:00]: And by the way, I also wanna make clear, right, we have been incredible-- Like we've taken a lot of advantage of the fact of all the open research that others have published, Right? And you mentioned, the Chinese labs, and we I think it's important that there's, from every country and every culture and background, including like Western companies like us, there's different models that come out that people can choose to trust. But I think we do have to give credit where credit's due, right? The incredible Chinese lab have done an amazing job at sharing their research, and we have definitely like been on the receiving end of taking advantage of that. So when you're on the receiving end of something coming to you, I think it's, you also have an obligation to give back.Swyx [00:30:39]: Do you have a favorite or underrated Chinese lab that you wanna shout out? Everyone shout outs DeepSeek.Chinese Labs, Zhipu, and PersistenceEiso Kant [00:30:44]: That's a good question.Swyx [00:30:45]: Moaan obviously for Therapsi. Yeah.Eiso Kant [00:30:48]: Yeah, look, I think, I think obviously everyone's been talking about Zhipu lately, with 5.2. I think what most people don't realize is when they started.Swyx [00:30:59]: Yeah.Eiso Kant [00:30:59]: Right? They started years before ChatGPT.Swyx [00:31:02]: They just rebranded. YeahEiso Kant [00:31:03]: And so, I've like, I remember how hard it was to work on these things Before the rest of the world got excited about it. And so I have an immense amount of respect for people, who were working on improving models when it wasn't the sexy thing to do, when believing in LLMs, was gonna get you ridiculed. I remember like back in 2016 when we were doing what we'd call, machine learning on code with some of these models. we would-- people would just laugh at us, like they'd be like, “This makes no sense. Like why are you wasting all these, like, millions of dollars on trying to figure this out?” And so I would say they're probably the one that, I think deserves a shout-out, not just because their latest model is very good, but because they fought to get here. And I think, I think every foundation model company it takes time to get here, right? It took us three years to get to the model that we're, that we're now gonna be releasing. and now the time in between the models is coming, is counted in weeks. It's no longer counted in months or years. But this stuff's hard. and if we can make it a little bit easier for the next person, like we should all do so. Because if we don't do so, we're, we've got a small window before models are really impacting recursive self-improvement to a level where catching up otherwise might become unfeasible. And we should try to, in that window, encourage as many labs or however we wanna call them, like to start. And so one of my currentEiso Kant [00:32:36]: Mission, but qualm is like I wanna encourage whoever is a researcher right now who thinks they can tackle this to go and leave and become my competitor.Eiso Kant [00:32:45]: Like start another foundation model company because I think we need it. I think otherwise we're not gonna be in the world where, I don't want to just be the fifth or the sixth company that wins. I wanna look at a world where there's lots of choice.Starting a Foundation Model CompanyVibhu [00:32:57]: What else do people not see in starting a foundation model? it's, there's a lot of compute, there's a lot of capital required, a lot of compute. You lay out model factory and how to do the training, but there's a lot there, right? That's,Eiso Kant [00:33:10]: Well, look, it's, I in turn-- this is an oversimplification, and I always asterisk it with that because it can land a little bit the wrong way in people's minds. But I think you can sum down, And I saw it, 95% of model building to just doing, you're just doing two things. You're improving data or you're improving compute efficiency. And I know that feels like an oversimplification for the incredible, like, Gifted and skilled work people do. But if you really look at it, like what are we doing? We are looking at data, we're generating new data, we're improving data. and the only way to do that is to look at the data, right? That's a big part of foundation model building. And on the other hand, we come up with these incredible breakthroughs in inference, in architecture, and new attention mechanisms. But what are they really doing? They're bringing compute efficiency. Now, we have definitely had some breakthroughs over the years that allow for more model capabilities. But at the limit, if you could train a large enough model, right, like, and you had infinite compute, we probably-- if you had infinite compute, you'd be at AGI probably already tomorrow.Eiso Kant [00:34:12]: Right? Like it's not. And so, and let me say that infinite compute with infinite ability of much faster networking because networking ends up being more of the bottleneck than compute. But, so I do think that's, those are the main things. And to just realize that this is engineering. I think it's become more obvious, but I think for quite a few years, people have held foundation model companies and researchers and others on this pedestal of like you're doing incredible magic or rocket science, or only like, Nobel laureate physicists can do this. And don't get me wrong, there are some really hard problems that need to be solved, but a lot of the work that all of us are doing on a day Is not sitting down trying to solve a math theorem. A lot of the work that we're doing is just really doing the basics right, writing good code, looking at data, improving it, running experiments, looking at plots, trying to see like, hey, trying to shape our intuitions. And a lot more people could be highly capable researchers. and I think that's, it feels far for people to do so. But I've seen in our own company, we've seen engineers become researchers because the model factory allowed them to be, have a much lower hurdle of running experiments and trying things. And one of the guys on our team who started as an engineer building our agents is a legit reinforcement learning researcher now, making real progress. and that happened in the span of like six months. that would've not been what I think most people assumed was possible, a couple of years ago.Swyx [00:35:46]: Yeah. I think one of the interesting moments is when you can self-host, like, if in a programming language, like if you can compile the language in the language, the equivalent is can you use your own tools, right? You have the pool CLI, you have your own models. presumably you're not only using your own models. There's no way. But like, what's that percentage over time?Laguna S, Persistence, and Behavioral GainsEiso Kant [00:36:10]: This is the first model that we're releasing that is starting to meaningfully contribute to our own work. It's not a it's not state-art model yet. Fable and other, they're, they're very capable models, but Laguna S Is really interesting. I'm gonna pull up the quote. Peng Ming, one of our heads of applied research, said something, last week as the model came out about 10 days ago, much better than we had hoped for or expected. And he said, I have the feeling that a lot of the gains in Laguna S come not from more intelligence, but more from different behavior, more verification, less taking things for granted, not declaring victory early, and being way more persistent. And to be honest, those are more predictive than raw intelligence for success in human also to some degree. And this was, he wrote me this on 5th of July on a Sunday, and it's been burned in my brain ever since because the Laguna S model, as you'll see it and why it does so well on benchmarks and why it does so well in using it on a day basis, is that it's just incredibly persistent. It reasons a lot. I do call that out. We have work to do on making it more efficient. We have to work to do on offering different reasoning modes. But this is the model that has been able to do things that I never thought it could do. A hundred eighteen billion 8B active model, which is not that large. It fits on a DGX Spark and still runs at, thirty, forty tokens a second on a Spark, is able to solve Erdős 397 independently. It's able to do complex programming tasks. It's able to. I asked it this morning to make me a Fi scanner without using any external libraries on my Mac, and it's, like, figuring out, like, the core WLAN API by really persistently trying to understand it without access to the internet. And more, I love vibe checking. I've probably spent eight to ten hours a day with this model for the last ten days.Eiso Kant [00:38:05]: I'm not exaggerating. I was on my eleven-hour flight yesterday. I spent ten hours reading trajectories and traces and, like, of the model.Eiso Kant [00:38:12]: And what I take away from it is exactly what Peng Ming said. We are gonna be able to squeeze so much more out of smaller models than I think we had imagined in the industry because, yes, there's intelligence and larger models are more intelligent. Like, no doubt about it. We should continue to scale up. but the behaviors of being really persistent, of being able to backtrack when you're wrong, of, like, understanding how to interact with your environment show us that we can get a lot more out of it. And this, for me, has created a bit of a Question in my mind the last couple of days. If you think about where we're using models today, right? We are using models, say, for knowledge work. Represents twenty-five percent of the global economy, twenty-five trillion dollars of work.Eiso Kant [00:39:00]: As we scale up models and they become more intelligent, we are excited about using them more and more for pushing the frontier of science.Small Models, Knowledge Work, and CommoditizationEiso Kant [00:39:08]: And if you look at the frontier of science, like true breakthroughs in science, they have been linked, they are linked to more intelligence in many places. Einstein figuring out general relativity is able to bring ideas together that other people would have not brought together. And I think one of the many dimensions of intelligence is the ability to do that, and it's something we clearly see that as models get larger and more capable, they're able to pull more ideas and threads together that a smaller model wouldn't be able to.Eiso Kant [00:39:36]: And we're starting to see examples of that in medicine and, like, in bio and other things. But if you think about the majority of knowledge work that we do, and it includes building software. I'm a software developer at heart first and foremost probably, although I probably can't say it that much anymore as I don't write production code in years, is that what makes us good is our persistence. It's our ability to encounter a problem and backtrack and say, “I need to go figure out this bug. I need to go research this. I need to go look at the documentation. I need to, like, try different, five different ways to see, like, if I can solve it.” But it is not necessarily bringing three ideas together from radically different fields. And so if we are now seeing, and I think Laguna S is an example, that we are able to make a relatively small model much more capable than I had definitely predicted or any previous, like, benchmarks had shown for any model remotely this size or even larger, At least on coding tasks, that it's because of the behaviors. And so now the question I have, and I don't have an answer, it is I know at the limit, so infinite model size, right, extremely large model, and the cost of that model is gonna be very expensive to run. We know this, right? So larger model ROI.Eiso Kant [00:40:52]: So I know that at the very limit, I'm not gonna use the world's largest model one day, quadrillion parameter, whatever crazy, like, scale we scale up, to do a basic coding task. Already today, I'm starting to size down for certain tasks.Eiso Kant [00:41:07]: So it means that there is an optimal. It means there's some curve that goes as we go up to model size for knowledge work, at some point we're at the peak, and after that, the return on investment of using a bigger model, just doesn't make sense.Eiso Kant [00:41:22]: Now, I think the question is, before I would have thought that peak was extremely very far away.Eiso Kant [00:41:30]: This model for me is the first sign that Maybe that peak is At a trillion, five trillion, ten trillion. Maybe we can just squeeze way more out of these models. I'm no longer thinking that we need two or three orders of magnitude on the largest models to be able to, solve knowledge work, the accounting, the legal, the code that we write. And so if that holds true, It is an argument for the commoditization of models. It's an argument that open source can win and, like, succeed in this world. And now it's of course a self-serving argument and it's a hopeful argument, but theoretically at the limit it works. We just have to go discover in the next couple of years of how much more we can squeeze out. Now, I do want to put a big asterisk. This does not mean I'm against scaling models. I think we ultimately only succeed if we scale our models as large as our competition. I do not like. I think we should not put our head in the sand and say we're gonna be king of open source small models. I think that's, It's a out. It's trying to be king of your own kingdom, but not realizing what the rest of the world's doing. All of us rather use a smarter, faster, more model. It's a sign of hope. And so I don't wanna overly state this is a good model. We have a long way to go to get to the state-art. But what hopefully people take away when they use this model is that the behaviors inside of it are what push it to be far more capable, less than necessarily the number of parameters.Pre-Training, Mid-Training, and RL Moving EarlierVibhu [00:43:03]: Is that mostly post-training? LikeEiso Kant [00:43:05]: YesVibhu [00:43:05]: Right.Eiso Kant [00:43:06]: It's entirely post-training.Vibhu [00:43:08]: Are we done improving anything on training? Is, like, training done?Eiso Kant [00:43:12]: No.Vibhu [00:43:12]: Okay.Eiso Kant [00:43:13]: SoVibhu [00:43:13]: I just wanted to cover training, and then we go post-trainingEiso Kant [00:43:15]: Training is not done. I mean, look, there's a part of training of just dealing with skill, right? Every new order of magnitude of model skill, you are going to get new things you gotta solve for. That'- but those are ultimately, engineering challenges.Eiso Kant [00:43:31]: I have a, I would say, a not commonly held opinion that reinforcement learning Will move earlier and earlier into training.Vibhu [00:43:42]: Yeah, training.Eiso Kant [00:43:44]: Not even training. Like training today, right, is, like if you look at - So we've been working on this for years already. and I think the best-- I think the first time we saw it out in public was the DeepSeek Zero paper. this is a year and a half ago, I think, if I recall correctly. where, you can Very early on in a model as it starts capable of being able to use language, et cetera, induce reasoning. and so the question that I have is like, we have this- we have the dataset that's the web. and the web, I think we could arguably say probably has The totality of humanity's knowledge somewhere encoded in different places. It's a huge variance degree of quality, from garbage data, and like once you look at training data, you really get humbled of like what the web is, to like, the most greatest scientific papers and best blog posts and like, best transcripts and whatnot.Eiso Kant [00:44:39]: And so now What we are trying to figure out, and have been doing a lot of work on, and it's a place where maybe not as open as we're on other things, but we will become more over time. we've been spending a couple of years really doing research on how can we turn the web into not just next token prediction, but into a way to teach the model to think earlier in its training. and I think there's a huge amount of gold to be found there. I think we are right now in, we've got some drugs in the industry. One of the drugs is distillation. Another drug is, more environments. Like, and they're great, and they make us feel good, and they make the models better, and like we're all addicted to them, and we'll use them, right? in various different ways. and but ultimately, I think we are still barely squeezing out of the web what we should be getting out of the web.Eiso Kant [00:45:33]: I think just next token prediction during training is not enough.Eiso Kant [00:45:36]: AndVibhu [00:45:38]: YeahEiso Kant [00:45:38]: I think we'll see some very interesting things still happen. and that RL in post-training to induce behaviors, to improve things, like I think - the whole world knows how to do this now. I think we're, we're scaling it up. Everyone is. But I wonder if we need to go as far as we're going today with environments. I'm not sure yetVibhu [00:46:01]: You mean we're going too far?Eiso Kant [00:46:02]: I'm, I'm not sure if the path to AGI is justVibhu [00:46:06]: Is more environmentEiso Kant [00:46:07]: More environments.Vibhu [00:46:08]: It seems like a never-ending, “Okay, I want instruction manual for this table, right? Am I gonna environment out building furniture? Or are we just gonna tail end like we need some general solution?”Eiso Kant [00:46:19]: I think there is, I think there's an ability to generalize more from the web. but I also am very encouraged, like when I look at Laguna S and, which is post-training is, well, is the big impact there. and I see like, oh, wait a second, just by making some of these behaviors much better, we're able to get so much more out of it. It just changes a little bit the way you think about intelligence.Vibhu [00:46:40]: Yeah. The analogy people draw often is the RL phase is where you don't learn as much new knowledge. You shiftEiso Kant [00:46:46]: Yeah.Vibhu [00:46:46]: Yeah. So, you shift distribution, and you can have it reason towards what you want. on your point about training, a lot of training is still just continue training in a domain, say medicine, then you do RL. So still justEiso Kant [00:47:00]: It's just better data, right? Like, I mean, training, ooh, I like how we invented this word. Like it's effectively just like,Vibhu [00:47:06]: Second phaseEiso Kant [00:47:07]: It's the second phase of training With like a really dumb way to do a curriculum. But like ultimately, what you'd want is a curriculum from token zero to token 30 whatever or 40 trillion tokens that really truly is the optimal curriculum for the model to learn. But training is essentially a stage curriculum on the web because we do not have to compute, And, effectively to try to ablate the perfect curriculum, right? And so I'm pretty sure that you'll start to see people talking soon about some other term, and there's two or - ‘cause now we do this, right? We talk stage two and stage three and stage four training and like. But ultimately, all we're doing is we're trying to assign a curriculum to the web data that we have to allow the model to learn better. I think at some point, as things get compute, as models get cheaper to run, as the next generations of compute, this will become more of a continuous spectrum. I also think the reason, by the way, you have training and like stage two and stage three is organizational, Right? It'- this is, I think, a thing where-- that we really try to avoid with the model factory is like Training exists because there's a training team now, right? There's people, or like people in training decide to focus on like a training effort. but what you really want is engineering and scale of experiments that allows for a much more continuous spectrum that you don't, you have infinite stages. Now, we're not there. Compute's not there. Organization design is not there for it yet. but I think we'll get there. we'll look back on a couple of years and be like, “Oh my God, it was so cute that we did our training data like this in such a like naïve way. Like we barely ordered it. We didn't really do a good job at likeCurriculum, Auto Research, and New ObjectivesVibhu [00:48:48]: The building that curriculum will get you that in the industry.Eiso Kant [00:48:51]: And I'll confirm that, when I talk to some researchers that this is a lot of the focus now is like how does training change and what is the next objective other than, next token prediction. I assume you don't have the answers, but you have some ideas.Vibhu [00:49:02]: We have some ideas. We're not ready to talk about it yet.Eiso Kant [00:49:05]: Yeah.Vibhu [00:49:05]: We've been working on them for years, and I think that's the one thing that's also like you asked earlier about, like what's not obvious about building a foundation model company is that you are constantly balancing the table stakes work, the recipe worksEiso Kant [00:49:19]: Yeah.Vibhu [00:49:19]: Versus like your, my crazyEiso Kant [00:49:22]: Pure researchVibhu [00:49:22]: Breakthrough.Eiso Kant [00:49:22]: Yeah.Vibhu [00:49:22]: Pure research and finding that balance and adjusting the percentage to it based on where you are in the race is really important.Eiso Kant [00:49:31]: I mean, so like, this is a nice way. I was gonna bring up auto research at some pointVibhu [00:49:35]: YesEiso Kant [00:49:35]: As another Andrej invention, or coinage, which is like, I honestly, like how many objective functions can there be, right? Like just try 1,000 of them, set it running, whatever.Vibhu [00:49:47]: Man, it's alsoEiso Kant [00:49:48]: Like what you're looking for. You're looking for loss curves like that, likeVibhu [00:49:51]: It's also a thing people take bets on, right? When you say more Neo labs, you're doing a version of we'll do foundation models, scale them up, next token predictors. A lot of other Neo labs that we see want to take a completely different approach, right? At some level, you're right. It's all, compute efficiency, and that's the net objective. But some are okay, different architecture, like vastly different amounts of compute spend. So some are different. They're not justEiso Kant [00:50:19]: YeahVibhu [00:50:19]: They're like, 99% not balancing, here's the vanilla and scale up. They're 99% on, here's novel research that'll change everything.Eiso Kant [00:50:27]: And I think, Luke, I think you. It depends when you started as well, right?Pure Research vs. Table StakesVibhu [00:50:30]: Yeah.Eiso Kant [00:50:30]: When we started, like the novel thing we did was reinforcement learning on code. No long- that's no longer novel by far, but we were like, - that's where we obsessed over when no one believed in RL. So you have to when you start the company, you have to have your own idea. You have to have something that's different that allows you to speed up, right? For us, it was RL to LLMs that later became common, like, Knowledge. But in the beginning, it wasn'tVibhu [00:50:53]: It's cool. this was like your original 2023 blogEiso Kant [00:50:57]: YeahVibhu [00:50:57]: Of purpose.Eiso Kant [00:50:58]: Yeah.Vibhu [00:50:59]: And like you do lay it all out here.Eiso Kant [00:51:01]: We laidVibhu [00:51:01]: The blog is pretty underrated, right? The whole RL on code was very early on.Eiso Kant [00:51:06]: Very early. And even we had to argue with people, like we say here things like to push beyond current capability, to train your own foundation model. We had to argue with people that it mattered that you had your own like, base model. you can fine-tune your way to success, right? major capabilities emerge from training a base model made accurate and useful during fine-tuning.Vibhu [00:51:23]: Which like, for perspective at the time, we knew closed models, OpenAI, Anthropic were huge. The open models we had were like Mistral 7B, a 30B, a 70B.Eiso Kant [00:51:35]: When weVibhu [00:51:35]: YeahEiso Kant [00:51:36]: The date on this thing is wrong. When we published this, it was April 2023. I think this was justVibhu [00:51:42]: YeahEiso Kant [00:51:42]: Happened on a migration, probably found it on archive.org.Vibhu [00:51:45]: Mistral.Eiso Kant [00:51:46]: Mistral had started, we started on the same month, right?Vibhu [00:51:49]: Yeah.Eiso Kant [00:51:49]: So this wasn't even, there was only, I think, Llama out at the timeVibhu [00:51:52]: SnellEiso Kant [00:51:52]: And that's it, right? And so, but I agree. I think we wan

This Week In Startups is made possible by: NetSuite https://NetSuite.ai/TWIST Squarespace https://squarespace.com/twist YSecurity https://YSecurity.io/TWIST Today's show: *America posted 400,000 farm jobs last year, and fewer than 1% got a single domestic applicant. Danny Bernstein of Reservoir believes it's time to start automating this backbreaking labor, like picking stone fruit in 110°F temperatures. So he built the world's first on-farm robotics incubator on 40 acres of California farmland. Here's why he argues that automating farms isn't just a new opportunity, it's a national security issue. PLUS Jason responded to Gal Shir's viral "AI beat me at design" tweet, got into a kerfuffle with Figma CEO Dylan Field, and wound up making a brand new J-Trade. AND we're talking about the Dept. of Education's new "do no harm" policy, the Brown University AI cheating chart that's blowing up social media, and Lon has fresh streaming recommendations in an all-new Off Duty. Guest: Danny Bernstein on X: https://x.com/bernsteind Reservoir Farms: https://reservoir.co/ Reservoir VC: https://reservoir.vc/ Relevant Links: Bonsai Robotics: https://bonsairobotics.ai/ Root AI acquired by AppHarvest: https://www.therobotreport.com/root-ai-acquired-by-appharvest-for-60m/ John Deere: https://www.deere.com/en-us/ Western Growers Association: https://www.wga.com/ Tanimura & Antle: https://www.taproduce.com/ Naturipe Berry Growers: https://www.naturipefarms.com/ Driscoll's: https://www.driscolls.com/ Taylor Farms: https://www.taylorfarms.com/ The Wonderful Company: https://www.wonderful.com/ Capital Factory: https://www.capitalfactory.com/ Gal Shir "quitting design" post: https://x.com/galshirart/status/2074854464729629060 Dylan Field response to Gal Shir: https://x.com/zoink/status/2075290218660298807 JCal response to Field and "J-Trade": https://x.com/Jason/status/2075481565115654305 Figma: https://www.figma.com/ Dept. of Education: "Do No Harm" policy announcement: https://www.ed.gov/about/news/press-release/us-department-of-education-issues-final-rule-hold-all-colleges-and-universities-accountable-low-earning-programs NPR coverage on Dept. of Education earnings test: https://www.npr.org/2026/06/30/nx-s1-5835631/turner-camhi-do-no-harm-college-loans Paul Graham "cheating chart" post: https://x.com/paulg/status/2075031014628311236 Off Duty Recommendations: "Lioness" on Paramount+: https://www.youtube.com/watch?v=jNRQ0PR4a8U "Mayor of Kingstown" on Paramount+: https://www.youtube.com/watch?v=VkQzvwxOp0s "Human Vapor" on Netflix: https://www.youtube.com/watch?v=7xe6dRKVAb8 "Sugar" on Apple TV+: https://www.youtube.com/watch?v=twvPGxuEOEA "Wind River" (now on Netflix): https://www.youtube.com/watch?v=CZgN0dpFoaE "Not Fade Away" by Peter Barton & Laurence Shames: https://www.amazon.com/Not-Fade-Away-Short-Lived/dp/1579546889 Timestamps: 0:00 Jason's ongoing World Tour 1:43 Danny Bernstein joins live from Reservoir Farms 3:38 What is "specialty crop agriculture" 5:15 Why strawberries are the "white whale" of AgTech 6:55 The labor crisis in farming 9:20 Why AgTech never scaled 9:51 NetSuite - For the first time, you can try NetSuite Next for free. If your revenues are at least in the seven figures, go to https://NetSuite.ai/TWIST 10:51 Inside Reservoir's business model 17:44 Agriculture as national security 20:09 Squarespace - Turn your idea into a beautiful website! Go to https://www.squarespace.com/twist for a free trial. When you're ready to launch, use offer code TWIST to save 10% off your first purchase of a website or domain. 22:15 Did AI beat a designer at design? 30:56 YSecurity - The on-demand security team for startups. Need enterprise-grade security without hiring a $400k CISO? YSecurity gives you 40+ expert engineers, matched to exactly what you need, by the hour, with your first six hours completely free. Go to https://YSecurity.io/TWIST 37:23 The "do no harm" college earnings test 43:27 Brown University students cheated on their midterms 52:22 Lon's streaming recommendations 59:46 Jason's new snake grabber   Subscribe to the TWiST500 newsletter: https://ticker.thisweekinstartups.com Check out the TWIST500: https://www.twist500.com Subscribe to This Week in Startups on Apple: https://rb.gy/v19fcp   Follow Lon: X: https://x.com/lons   Follow Alex: X: https://x.com/alex LinkedIn: ⁠https://www.linkedin.com/in/alexwilhelm   Follow Jason: X: https://twitter.com/Jason LinkedIn: https://www.linkedin.com/in/jasoncalacanis   Check out all our partner offers: https://partners.launch.co/   Great TWIST interviews: Will Guidara, Eoghan McCabe, Steve Huffman, Brian Chesky, Bob Moesta, Aaron Levie, Sophia Amoruso, Reid Hoffman, Frank Slootman, Billy McFarland   Check out Jason's suite of newsletters: https://substack.com/@calacanis   Follow TWiST: Twitter: https://twitter.com/TWiStartups YouTube: https://www.youtube.com/thisweekin Instagram: https://www.instagram.com/thisweekinstartups TikTok: https://www.tiktok.com/@thisweekinstartups Substack: https://twistartups.substack.com

This is a preview | For full audio and show notes plus extras, subscribe via https://patreon.com/newmodels or https://newmodels.substack.com _ When Enlightenment-era scholars taxonomized humankind, they chose homo sapiens, in part because of what they observed as a uniquely human aptitude for “taste.” [“Taste” comes from the Latin taxare (to handle, to assess) – an intensive form of tangere (to touch). It is a term that via vulgar Latin blended with gustare (to taste, to try) to express the act of appraising something by physically handling or even consuming it. In proper Latin, there is the related term sapere, which carries a double meaning of “to taste” and “to be wise.” To taste, in the sense of sapere, is to possess wisdom through tangible experience.] It's remarkable that in recent years, and intensely in recent months, the idea of “taste”—“having taste,” “deploying taste,” “taste agents,” “integrating a taste layer,” “taste as a core skill,” “taste as moat”—has become an ultra-present concern in tech circles. Isn't having taste (wherever one falls along the taste spectrum) an inherent quality of being human? To be sure, there's already a lot of writing on this phenomenon: “Tasteslop” by NEMESIS‘s Emily Segal and “Why Tech Bros Are Now Obsessed With Taste” by Kyle Chayka for the New Yorker being among the best takes. The most confounding, in our opinion being Y-Combinator cofounder Paul Graham's “Taste for Makers” (Feb 2002, but highly cited this year). There is also “Against Taste,” by Will Manidis, which makes some good points (patrons used to fund art for a higher power or at least public display whereas contemporary collectors tend to buy art for their own private use) but in its theory-of-everything aspiration, feels LLM-ish and contextually ahistorical. So in pure “Content Today” form, we are adding to the pile-on with our own fashionably late, probably factually botched but definitely human, free-associative conversation about taste.

If you've ever been to PyCon, you know one of the best parts of the expo hall is Startup Row, a stretch of booths where early-stage companies built on Python show off what they're creating. But only attendees get to walk that lane, so let's bring it to everyone. In this episode, we stroll down Startup Row together. We kick things off with the organizers, Jason and Shay, who share the program's origin story going back to Paul Graham and the PSF, plus some surprising stats, including two unicorns among the alumni. Then we meet five startups: Tetrix, bringing AI to institutional investing in private markets. Arcjet, security that lives inside your app as an SDK. Phemeral.dev, serverless hosting built for Python web apps. CapiscIO, an identity and authority layer for AI agents. And Pixeltable, a multimodal database from Marcel Kornacker, co-creator of Apache Parquet. See if you can spot the theme running through them all. Let's go for a walk. Episode sponsors AgentField AI Talk Python Courses Links from the show Guests Naunidh Bhalla: linkedin.com Grant Gittes: linkedin.com Marcel Kornacker: linkedin.com Beon de Nood: linkedin.com Chinmaya Joshi: linkedin.com David Mytton: linkedin.com Shea Tate-Di Donna: linkedin.com Jason Rowley: linkedin.com Azul Garza: github.com Renée Rosillo: linkedin.com Tetrix: tetrix.co Tetrix Jobs: tetrix.co Arcjet: arcjet.com Pixeltable: pixeltable.com Phemeral.dev: phemeral.dev CapiscIO: capisc.io Episode #551 deep-dive: talkpython.fm/551 Episode transcripts: talkpython.fm Theme Song: Developer Rap

⁠Geoff Ralston⁠⁠ built Rocket Mail before Yahoo Mail existed, built Lala before Spotify worked in America, sat in Steve Jobs' living room to get an acquisition approved, and ran Y Combinator as President. Now he says the most important thing he's ever worked on is making sure AI doesn't kill us.In this episode with ⁠⁠Kyriakos Eleftheriou⁠⁠, Geoff speaks about YC, SAIF, the Yahoo acquisition, and the top learnings after decades in silicon valley00:00 Introduction03:45 The $70K bet that built Sand Hill Road06:07 Early to every wave06:59 "The most important thing I've ever worked on"08:18 Timing is luck. Here's how to get lucky anyway10:46 Quitting HP the day he saw the Mosaic browser15:39 Paul Graham's "packets of cash" acquisition rule19:35 The Yahoo revolt that almost killed the deal33:45 Zuck killed Facebook Music in one sentence39:05 Why Steve Jobs offered him a job — and why he said no45:40 Running Y Combinator50:37 Steve Jobs' real superpower54:06 Why AI is more dangerous than people think01:01:15 Peter Thiel is wrong about competition01:03:31 Marc Andreessen is wrong about introspection--Full podcast and writeup → Geoff Ralston on early YC, AI risk, and why Thiel is wrong about competition

Send us Fan MailWhen GPT-4 rendered Ali Dastjerdi's product obsolete overnight, most founders would have doubled down. He paused everything, rebuilt from first principles, and landed in a market managing $500 billion in AUM. The lesson isn't about pivoting - it's about building for the model that doesn't exist yet, so every new AI release accelerates your business instead of threatening it.What You Will LearnHow to know when to stop fighting your existing product and rebuild from zeroWhy the investor who believed first — not the largest fund — is the one worth chasingWhat it means to build a company that cheers at AI announcements instead of fearing themHow Ali's own customer led his Series A and what that signals about product-market fitWhy raising money is about finding people who already believe — not convincing the scepticalAbout the GuestAli Dastjerdi is co-founder and CEO of Raylu, an AI-native deal flow platform now serving over 60 funds managing approximately $500 billion in combined AUM, including four of the world's top 25 private equity firms. Before founding Raylu, Ali spent four years as an investor at Insight Partners. Raylu builds AI agents that automate the full lifecycle of proprietary deal sourcing for private market funds. Connect with Ali on LinkedIn and follow RayluTimestamps 00:00 — Three friends, one WeWork office, and no product-market fit 03:19 — The night GPT-4 made everything they built obsolete 06:25 — 12 months of pivots and the moment the team started breaking 09:05 — How a Paul Graham lecture led them to the $500B opportunity 11:25 — What Raylu actually does and why private markets need it now 15:39 — How their own customer led their Series A 17:46 — What Highland X was actually betting onAbout the GuestAli's LinkedInRaylu AI WebsiteConnect with HinaHina's WebsiteHina's LinkedInHina's InstagramHina's Youtube Channel Hina's Email Production Credit: Produced by @the32collective_ / https://www.the32collective.co/

Raccontiamo un grande classico di Paul Graham del 2004, il "Python Paradox". Perché un'azienda che lavora in Java avrebbe dovuto assumere un programmatore Python nel 2004?Blog post originale: The Python Paradox

Den, i dag, smått legendariska startupfabriken Y Combinator grundades i Silicon Valley 2005 av bland annat Paul Graham och Jessica Livingston, som nyligen besökte Stockholm. Björn Jeffery berättar om när bolaget delade lokal med ett robotföretag. Henning Eklund ifrågasätter hur framgångsrika de egentligen är. Sophia Sinclair menar att de kanske mest ger startups en bra förberedelse för nästa steg.

From seventeen to thousands, our guest joins us to tell the story of how a young man from a small town established a thriving revival church in a major urban center. Listen as he discusses the impact of multiculturalism, church streaming, and consistent prayer.   #KingdomSpeak #Podcast  #Revival

There is a long-standing stigma in venture capital that debt is a "company killer." This week on The Data Minute, Peter Walker sits down with Marshall Hawks, former SVB expert and author of “Venture Debt Deals,” to debunk the myths and explain why debt is often the smartest addition to a founder's equity mix.Marshall breaks down the tactical reality of how these deals actually get done, from the "sniff test" lenders perform during office visits to the critical differences between venture banks and private credit funds. He explains how founders can use debt to survive 15-year exit timelines while minimizing dilution, and shares the specific red flags that indicate a startup is becoming over-leveraged.Plus, Marshall offers a rare look at the "workout groups" that step in when things go wrong and explains why a company's General Counsel might not be the right person to lead a debt negotiation. Whether you are an early-stage founder or a late-stage operator, this episode is a definitive guide to capitalizing your business in a shifting market.Subscribe to Carta's weekly Data Minute newsletter: https://carta.com/subscribe/data-newsletter-sign-up/Explore interactive startup and VC data, with Carta's Data Desk: https://carta.com/data-desk/Chapters:00:16 – Intro: Marshall Hawks and the Venture Debt stigma01:10 – Why Marshall wrote "Venture Debt Deals"03:26 – Addressing the Paul Graham view: Is debt dangerous?06:40 – When (and when NOT) to touch venture debt09:14 – The "Insurance" Play: Why many founders never draw the capital10:48 – Venture Banks vs. Private Credit Funds13:00 – Understanding draw periods and interest-only terms17:34 – Why your lender wants to visit your office (The Sniff Test)22:28 – The market after March 2023: Life after SVB26:09 – The "Workout Group": What happens when things go sideways? 30:31 – Green flags: How to diligence your lending partner33:50 – The legal process: Why GC's need outside support37:13 – Hidden costs: Why the company pays everyone's legal fees42:31 – Using debt to survive 15-year exit timelines44:49 – Red flags: Debt service vs. opex ratios48:06 – Final advice: Fundraising is not successThis presentation contains general information only and eShares, Inc. dba Carta, Inc. (“Carta”) is not, by means of this publication, rendering accounting, business, financial, investment, legal, tax, or other professional advice or services, and is for informational purposes only.  This presentation is not a substitute for such professional advice or services nor should it be used as a basis for any decision or action that may affect your business or interests. © 2026 eShares, Inc., dba Carta, Inc. All rights reserved.

durée : 00:10:44 - L'Invité(e) des Matins - par : Guillaume Erner - Sam Altman, dans le contexte du développement de sa première start-up, Loopt, noue un lien étroit avec celui qui deviendra son mentor, Paul Graham, une figure méconnue mais centrale du monde de la tech. Ce dernier confie d'ailleurs les rênes de son entreprise, Y Combinator, à Sam Altman, en 2014. - invités : Olivier Alexandre Sociologue et directeur adjoint du Centre Internet et Société du CNRS.

durée : 00:10:44 - L'Invité(e) des Matins - par : Guillaume Erner - Sam Altman, dans le contexte du développement de sa première start-up, Loopt, noue un lien étroit avec celui qui deviendra son mentor, Paul Graham, une figure méconnue mais centrale du monde de la tech. Ce dernier confie d'ailleurs les rênes de son entreprise, Y Combinator, à Sam Altman, en 2014. - invités : Olivier Alexandre Sociologue et directeur adjoint du Centre Internet et Société du CNRS.

durée : 00:10:44 - Cultures monde - par : Guillaume Erner - Sam Altman, dans le contexte du développement de sa première start-up, Loopt, noue un lien étroit avec celui qui deviendra son mentor, Paul Graham, une figure méconnue mais centrale du monde de la tech. Ce dernier confie d'ailleurs les rênes de son entreprise, Y Combinator, à Sam Altman, en 2014. - réalisation : Juliette Devaux, Alice Deschamps - invités : Olivier Alexandre Sociologue et directeur adjoint du Centre Internet et Société du CNRS. Vous aimez ce podcast ? Pour écouter tous les épisodes sans limite, rendez-vous sur Radio France

Did you like the episode? Send me a text and let me know!! Maker Schedules for Solo Founders: How to Scale Without BurnoutAre you a solo founder frantically laying down train tracks while the train is speeding right behind you? In this episode, we tear up generic "hustle culture" to build a customized, energy-driven time management engine specifically for the solopreneur. Learn how to ditch the busywork, protect your deep creative focus, and scale your business without breaking yourself in the process.

Veterinary medicine has long relied on "snapshots" of health taken during clinic visits, but what happens during the months in between? This week, Shawn Wilkie and Dr. Ivan Zak speak with Eric Humbert, Head of Science & AI at Invoxia, about how wearable technology is helping veterinarians monitor pets more continuously and more objectively. Eric explains how the Biotracker uses advanced signal processing and AI to track heart rate, respiratory rate, activity, and recovery trend outside the clinic — and how the Biotrack veterinary platform gives clinics a dedicated dashboard to monitor all their patients remotely. From earlier detection of chronic disease deterioration to post-op follow-up and treatment monitoring, the episode looks at how remote monitoring could support more proactive care while also opening new opportunities for client engagement and recurring revenue. The episode offers a practical look at where wearable data may fit into real veterinary workflows, what still needs to be solved, and why continuous monitoring could become an important part of the future of care. Learn more about Invoxia. Discover more at Biotrack Invoxia. Eric recommends "Hackers & Painters: Big Ideas from the Computer Age" by Paul Graham for its stimulating ideas on technology, independent thinking, and the power of startups.

Guest post by Julian Lighton author of Navigating Your Next: Discover the Career You Want and the Path to Get There. There has been much debate since Paul Graham's provocative 2024 article coining the term 'Founder Mode' about its benefits vs 'Manager Mode' and how they impact success in the scaling transition. My own view is that a more balanced mindset and skills set is required – an entrepreneurial leadership approach. Navigating successful growth I've spent over twenty years working with over a hundred VC and PE backed early stage and scaling businesses in both tech and other industries, including some of the most successful businesses in their categories, such as Gainsight, Renaissance Learning, Corsair Gaming and SnapNurse. Based on this experience, the transition between founder and scaling is the single most important test of a CEO's adaptability and leadership. Most leaders underestimate how personally challenging it is and how much they need to change as they climb the staircases of growth. Here are five changes in mindset and skillset that are key for successful growth and building companies that are built to last: From identity as founder to identity as leader: From "I do it" and "I decide" to "We do it" and "We decide." The founder's initial motivation—often deeply personal—must now become collective. Why does the team care? Why should they go the extra mile? If you can help your people answer "Why do we want this?" (know why) and help them see themselves in the company's success (care why), you unlock discretionary effort and loyalty that no compensation package alone can buy. The CEO's job becomes less about being the smartest person in the room and more about building teams that can make great decisions without you, without constant intervention. The focus shifts to hiring and retaining great people and letting them be responsible, setting very clear direction (the what not the how), building culture, and putting in place the right incentives and feedback loops. It's about shifting from heroics to scalability. From implicit to explicit: Scaling requires clarity. Being implicit – carrying everything in the CEO's head does not scale. Scaling requires clarity about roles and responsibilities; plans that everyone can understand and follow – communicating simply and exactly what is required, why it matters, and who will do what; governance and decision making; metrics and more important than all of that clear culture. A study by Columbia Business School found that when you add more than 20% new joiners, priorities, and values get diluted and team cohesion and psychological safety can break down. Without clear definition and communication about who is responsible for what, teams become inefficient, drop balls, and experience internal friction. This undermines performance and accountability, trust and collaboration erode and performance suffers. From who got you here, to who will get you there: People who worked when your organization was smaller often break when you try to scale. No amount of ambition or capital can compensate for the wrong team members with the wrong skills or a lack of alignment. Scaling exposes weaknesses in team leadership, skills, and coordination. Ask yourself: do you have the right people in the right seats for this stage of growth? Are your team leaders and teams aligned, and does everyone understand the goals and the urgency? Invest in hiring and developing talent density in teams. Ensure the team's incentives (care why) and goals are in sync with the business's direction (know why). From measuring, to measuring what matters: What really drives success at what stage of growth? Ruthless prioritization is essential. In scaling organizations, it's common for teams to focus on metrics that were appropriate for the previous stage (staircase) of growth e.g. product adoption vs revenue; new business vs repeat business; revenue vs profits. But without visible progress markers and regular, discipl...

AI founders are finally turning to paid marketing… and the internet is losing its mind!This week, James and Daniel unpack the debate sparked by Andrew Chen and Bill Gurley, and question whether paid marketing really kills creativity. They also explore why, despite 25+ years of data, there's still no clear startup playbook and why that might never change.Plus: the shift from hiring people to hiring tech, why AI could reward leaders like Mark Zuckerberg, and a few bold takes from Paul Graham.It's the funniest era of tech, where everything is easier to build, harder to grow, and somehow… ads still work.STAY CONNECTEDJAMES on Twitter & Linkedin – /jamesborowDANIEL on LinkedIn, Instagram, TikTok – /danieldruger

Join Kyle, Nader, Vibhu, and swyx live at NVIDIA GTC next week!Now that AIE Europe tix are ~sold out, our attention turns to Miami and World's Fair!The definitive AI Accelerator chip company has more than 10xed this AI Summer:And is now a $4.4 trillion megacorp… that is somehow still moving like a startup. We are blessed to have a unique relationship with our first ever NVIDIA guests: Kyle Kranen who gave a great inference keynote at the first World's Fair and is one of the leading architects of NVIDIA Dynamo (a Datacenter scale inference framework supporting SGLang, TRT-LLM, vLLM), and Nader Khalil, a friend of swyx from our days in Celo in The Arena, who has been drawing developers at GTC since before they were even a glimmer in the eye of NVIDIA:Nader discusses how NVIDIA Brev has drastically reduced the barriers to entry for developers to get a top of the line GPU up and running, and Kyle explains NVIDIA Dynamo as a data center scale inference engine that optimizes serving by scaling out, leveraging techniques like prefill/decode disaggregation, scheduling, and Kubernetes-based orchestration, framed around cost, latency, and quality tradeoffs. We also dive into Jensen's “SOL” (Speed of Light) first-principles urgency concept, long-context limits and model/hardware co-design, internal model APIs (https://build.nvidia.com), and upcoming Dynamo and agent sessions at GTC.Full Video pod on YouTubeTimestamps00:00 Agent Security Basics00:39 Podcast Welcome and Guests07:19 Acquisition and DevEx Shift13:48 SOL Culture and Dynamo Setup27:38 Why Scale Out Wins29:02 Scale Up Limits Explained30:24 From Laptop to Multi Node33:07 Cost Quality Latency Tradeoffs38:42 Disaggregation Prefill vs Decode41:05 Kubernetes Scaling with Grove43:20 Context Length and Co Design57:34 Security Meets Agents58:01 Agent Permissions Model59:10 Build Nvidia Inference Gateway01:01:52 Hackathons And Autonomy Dreams01:10:26 Local GPUs And Scaling Inference01:15:31 Long Running Agents And SF ReflectionsTranscriptAgent Security BasicsNader: Agents can do three things. They can access your files, they can access the internet, and then now they can write custom code and execute it. You literally only let an agent do two of those three things. If you can access your files and you can write custom code, you don't want internet access because that's one to see full vulnerability, right?If you have access to internet and your file system, you should know the full scope of what that agent's capable of doing. Otherwise, now we can get injected or something that can happen. And so that's a lot of what we've been thinking about is like, you know, how do we both enable this because it's clearly the future.But then also, you know, what, what are these enforcement points that we can start to like protect?swyx: All right.Podcast Welcome and Guestsswyx: Welcome to the Lean Space podcast in the Chromo studio. Welcome to all the guests here. Uh, we are back with our guest host Viu. Welcome. Good to have you back. And our friends, uh, Netter and Kyle from Nvidia. Welcome.Kyle: Yeah, thanks for having us.swyx: Yeah, thank you. Actually, I don't even know your titles.Uh, I know you're like architect something of Dynamo.Kyle: Yeah. I, I'm one of the engineering leaders [00:01:00] and a architects of Dynamo.swyx: And you're director of something and developers, developer tech.Nader: Yeah.swyx: You're the developers, developers, developers guy at nvidia,Nader: open source agent marketing, brev,swyx: and likeNader: Devrel tools and stuff.swyx: Yeah. BeenNader: the focus.swyx: And we're, we're kind of recording this ahead of Nvidia, GTC, which is coming to town, uh, again, uh, or taking over town, uh, which, uh, which we'll all be at. Um, and we'll talk a little bit about your sessions and stuff. Yeah.Nader: We're super excited for it.GTC Booth Stunt Storiesswyx: One of my favorite memories for Nader, like you always do like marketing stunts and like while you were at Rev, you like had this surfboard that you like, went down to GTC with and like, NA Nvidia apparently, like did so much that they bought you.Like what, what was that like? What was that?Nader: Yeah. Yeah, we, we, um. Our logo was a chaka. We, we, uh, we were always just kind of like trying to keep true to who we were. I think, you know, some stuff, startups, you're like trying to pretend that you're a bigger, more mature company than you are. And it was actually Evan Conrad from SF Compute who was just like, you guys are like previousswyx: guest.Yeah.Nader: Amazing. Oh, really? Amazing. Yeah. He was just like, guys, you're two dudes in the room. Why are you [00:02:00] pretending that you're not? Uh, and so then we were like, okay, let's make the logo a shaka. We brought surfboards to our booth to GTC and the energy was great. Yeah. Some palm trees too. They,Kyle: they actually poked out over like the, the walls so you could, you could see the bread booth.Oh, that's so funny. AndNader: no one else,Kyle: just from very far away.Nader: Oh, so you remember it backKyle: then? Yeah I remember it pre-acquisition. I was like, oh, those guys look cool,Nader: dude. That makes sense. ‘cause uh, we, so we signed up really last minute, and so we had the last booth. It was all the way in the corner. And so I was, I was worried that no one was gonna come.So that's why we had like the palm trees. We really came in with the surfboards. We even had one of our investors bring her dog and then she was just like walking the dog around to try to like, bring energy towards our booth. Yeah.swyx: Steph.Kyle: Yeah. Yeah, she's the best,swyx: you know, as a conference organizer, I love that.Right? Like, it's like everyone who sponsors a conference comes, does their booth. They're like, we are changing the future of ai or something, some generic b******t and like, no, like actually try to stand out, make it fun, right? And people still remember it after three years.Nader: Yeah. Yeah. You know what's so funny?I'll, I'll send, I'll give you this clip if you wanna, if you wanna add it [00:03:00] in, but, uh, my wife was at the time fiance, she was in medical school and she came to help us. ‘cause it was like a big moment for us. And so we, we bought this cricket, it's like a vinyl, like a vinyl, uh, printer. ‘cause like, how else are we gonna label the surfboard?So, we got a surfboard, luckily was able to purchase that on the company card. We got a cricket and it was just like fine tuning for enterprises or something like that, that we put on the. On the surfboard and it's 1:00 AM the day before we go to GTC. She's helping me put these like vinyl stickers on.And she goes, you son of, she's like, if you pull this off, you son of a b***h. And so, uh, right. Pretty much after the acquisition, I stitched that with the mag music acquisition. I sent it to our family group chat. Ohswyx: Yeah. No, well, she, she made a good choice there. Was that like basically the origin story for Launchable is that we, it was, and maybe we should explain what Brev is andNader: Yeah.Yeah. Uh, I mean, brev is just, it's a developer tool that makes it really easy to get a GPU. So we connect a bunch of different GPU sources. So the basics of it is like, how quickly can we SSH you into a G, into a GPU and whenever we would talk to users, they wanted A GPU. They wanted an A 100. And if you go to like any cloud [00:04:00] provisioning page, usually it's like three pages of forms or in the forms somewhere there's a dropdown.And in the dropdown there's some weird code that you know to translate to an A 100. And I remember just thinking like. Every time someone says they want an A 100, like the piece of text that they're telling me that they want is like, stuffed away in the corner. Yeah. And so we were like, what if the biggest piece of text was what the user's asking for?And so when you go to Brev, it's just big GPU chips with the type that you want withswyx: beautiful animations that you worked on pre, like pre you can, like, now you can just prompt it. But back in the day. Yeah. Yeah. Those were handcraft, handcrafted artisanal code.Nader: Yeah. I was actually really proud of that because, uh, it was an, i I made it in Figma.Yeah. And then I found, I was like really struggling to figure out how to turn it from like Figma to react. So what it actually is, is just an SVG and I, I have all the styles and so when you change the chip, whether it's like active or not it changes the SVG code and that somehow like renders like, looks like it's animating, but it, we just had the transition slow, but it's just like the, a JavaScript function to change the like underlying SVG.Yeah. And that was how I ended up like figuring out how to move it from from Figma. But yeah, that's Art Artisan. [00:05:00]Kyle: Speaking of marketing stunts though, he actually used those SVGs. Or kind of use those SVGs to make these cards.Nader: Oh yeah. LikeKyle: a GPU gift card Yes. That he handed out everywhere. That was actually my first impression of thatNader: one.Yeah,swyx: yeah, yeah.Nader: Yeah.swyx: I think I still have one of them.Nader: They look great.Kyle: Yeah.Nader: I have a ton of them still actually in our garage, which just, they don't have labels. We should honestly like bring, bring them back. But, um, I found this old printing press here, actually just around the corner on Ven ness. And it's a third generation San Francisco shop.And so I come in an excited startup founder trying to like, and they just have this crazy old machinery and I'm in awe. ‘cause the the whole building is so physical. Like you're seeing these machines, they have like pedals to like move these saws and whatever. I don't know what this machinery is, but I saw all three generations.Like there's like the grandpa, the father and the son, and the son was like, around my age. Well,swyx: it's like a holy, holy trinity.Nader: It's funny because we, so I just took the same SVG and we just like printed it and it's foil printing, so they make a a, a mold. That's like an inverse of like the A 100 and then they put the foil on it [00:06:00] and then they press it into the paper.And I remember once we got them, he was like, Hey, don't forget about us. You know, I guess like early Apple and Cisco's first business cards were all made there. And so he was like, yeah, we, we get like the startup businesses but then as they mature, they kind of go somewhere else. And so I actually, I think we were talking with marketing about like using them for some, we should go back and make some cards.swyx: Yeah, yeah, yeah. You know, I remember, you know, as a very, very small breadth investor, I was like, why are we spending time like, doing these like stunts for GPUs? Like, you know, I think like as a, you know, typical like cloud hard hardware person, you go into an AWS you pick like T five X xl, whatever, and it's just like from a list and you look at the specs like, why animate this GP?And, and I, I do think like it just shows the level of care that goes throughout birth and Yeah. And now, and also the, and,Nader: and Nvidia. I think that's what the, the thing that struck me most when we first came in was like the amount of passion that everyone has. Like, I think, um, you know, you talk to, you talk to Kyle, you talk to, like, every VP that I've met at Nvidia goes so close to the metal.Like, I remember it was almost a year ago, and like my VP asked me, he's like, Hey, [00:07:00] what's cursor? And like, are you using it? And if so, why? Surprised at this, and he downloaded Cursor and he was asking me to help him like, use it. And I thought that was, uh, or like, just show him what he, you know, why we were using it.And so, the amount of care that I think everyone has and the passion, appreciate, passion and appreciation for the moment. Right. This is a very unique time. So it's really cool to see everyone really like, uh, appreciate that.swyx: Yeah.Acquisition and DevEx Shiftswyx: One thing I wanted to do before we move over to sort of like research topics and, uh, the, the stuff that Kyle's working on is just tell the story of the acquisition, right?Like, not many people have been, been through an acquisition with Nvidia. What's it like? Uh, what, yeah, just anything you'd like to say.Nader: It's a crazy experience. I think, uh, you know, we were the thing that was the most exciting for us was. Our goal was just to make it easier for developers.We wanted to find access to GPUs, make it easier to do that. And then all, oh, actually your question about launchable. So launchable was just make one click exper, like one click deploys for any software on top of the GPU. Mm-hmm. And so what we really liked about Nvidia was that it felt like we just got a lot more resources to do all of that.I think, uh, you [00:08:00] know, NVIDIA's goal is to make things as easy for developers as possible. So there was a really nice like synergy there. I think that, you know, when it comes to like an acquisition, I think the amount that the soul of the products align, I think is gonna be. Is going speak to the success of the acquisition.Yeah. And so it in many ways feels like we're home. This is a really great outcome for us. Like we you know, I love brev.nvidia.com. Like you should, you should use it's, it's theKyle: front page for GPUs.Nader: Yeah. Yeah. If you want GP views,Kyle: you go there, getswyx: it there, and it's like internally is growing very quickly.I, I don't remember You said some stats there.Nader: Yeah, yeah, yeah. It's, uh, I, I wish I had the exact numbers, but like internally, externally, it's been growing really quickly. We've been working with a bunch of partners with a bunch of different customers and ISVs, if you have a solution that you want someone that runs on the GPU and you want people to use it quickly, we can bundle it up, uh, in a launchable and make it a one click run.If you're doing things and you want just like a sandbox or something to run on, right. Like open claw. Huge moment. Super exciting. Our, uh, and we'll talk into it more, but. You know, internally, people wanna run this, and you, we know we have to be really careful from the security implications. Do we let this run on the corporate network?Security's guidance was, Hey, [00:09:00] run this on breath, it's in, you know, it's, it's, it's a vm, it's sitting in the cloud, it's off the corporate network. It's isolated. And so that's been our stance internally and externally about how to even run something like open call while we figure out how to run these things securely.But yeah,swyx: I think there's also like, you almost like we're the right team at the right time when Nvidia is starting to invest a lot more in developer experience or whatever you call it. Yeah. Uh, UX or I don't know what you call it, like software. Like obviously NVIDIA is always invested in software, but like, there's like, this is like a different audience.Yeah. It's aNader: widerKyle: developer base.swyx: Yeah. Right.Nader: Yeah. Yeah. You know, it's funny, it's like, it's not, uh,swyx: so like, what, what is it called internally? What, what is this that people should be aware that is going on there?Nader: Uh, what, like developer experienceswyx: or, yeah, yeah. Is it's called just developer experience or is there like a broader strategy hereNader: in Nvidia?Um, Nvidia always wants to make a good developer experience. The thing is and a lot of the technology is just really complicated. Like, it's not, it's uh, you know, I think, um. The thing that's been really growing or the AI's growing is having a huge moment, not [00:10:00] because like, let's say data scientists in 2018, were quiet then and are much louder now.The pie is com, right? There's a whole bunch of new audiences. My mom's wondering what she's doing. My sister's learned, like taught herself how to code. Like the, um, you know, I, I actually think just generally AI's a big equalizer and you're seeing a more like technologically literate society, I guess.Like everyone's, everyone's learning how to code. Uh, there isn't really an excuse for that. And so building a good UX means that you really understand who your end user is. And when your end user becomes such a wide, uh, variety of people, then you have to almost like reinvent the practice, right? Yeah. You haveKyle: to, and actually build more developer ux, right?Because the, there are tiers of developer base that were added. You know, the, the hackers that are building on top of open claw, right? For example, have never used gpu. They don't know what kuda is. They, they, they just want to run something.Nader: Yeah.Kyle: You need new UX that is not just. Hey, you know, how do you program something in Cuda and run it?And then, and then we built, you know, like when Deep Learning was getting big, we built, we built Torch and, and, but so recently the amount of like [00:11:00] layers that are added to that developer stack has just exploded because AI has become ubiquitous. Everyone's using it in different ways. Yeah. It'sNader: moving fast in every direction.Vertical, horizontal.Vibhu: Yeah. You guys, you even take it down to hardware, like the DGX Spark, you know, it's, it's basically the same system as just throwing it up on big GPU cluster.Nader: Yeah, yeah, yeah. It's amazing. Blackwell.swyx: Yeah. Uh, we saw the preview at the last year's GTC and that was one of the better performing, uh, videos so far, and video coverage so far.Awesome. This will beat it. Um,Nader: that wasswyx: actually, we have fingersNader: crossed. Yeah.DGX Spark and Remote AccessNader: Even when Grace Blackwell or when, um, uh, DGX Spark was first coming out getting to be involved in that from the beginning of the developer experience. And it just comes back to what youswyx: were involved.Nader: Yeah. St. St.swyx: Mars.Nader: Yeah. Yeah. I mean from, it was just like, I, I got an email, we just got thrown into the loop and suddenly yeah, I, it was actually really funny ‘cause I'm still pretty fresh from the acquisition and I'm, I'm getting an email from a bunch of the engineering VPs about like, the new hardware, GPU chip, like we're, or not chip, but just GPU system that we're putting out.And I'm like, okay, cool. Matters. Now involved with this for the ux, I'm like. What am I gonna do [00:12:00] here? So, I remember the first meeting, I was just like kind of quiet as I was hearing engineering VPs talk about what this box could be, what it could do, how we should use it. And I remember, uh, one of the first ideas that people were idea was like, oh, the first thing that it was like, I think a quote was like, the first thing someone's gonna wanna do with this is get two of them and run a Kubernetes cluster on top of them.And I was like, oh, I think I know why I'm here. I was like, the first thing we're doing is easy. SSH into the machine. And then, and you know, just kind of like scoping it down of like, once you can do that every, you, like the person who wants to run a Kubernetes cluster onto Sparks has a higher propensity for pain, then, then you know someone who buys it and wants to run open Claw right now, right?If you can make sure that that's as effortless as possible, then the rest becomes easy. So there's a tool called Nvidia Sync. It just makes the SSH connection really simple. So, you know, if you think about it like. If you have a Mac, uh, or a PC or whatever, if you have a laptop and you buy this GPU and you want to use it, you should be able to use it like it's A-A-G-P-U in the cloud, right?Um, but there's all this friction of like, how do you actually get into that? That's part of [00:13:00] Revs value proposition is just, you know, there's a CLI that wraps SSH and makes it simple. And so our goal is just get you into that machine really easily. And one thing we just launched at CES, it's in, it's still in like early access.We're ironing out some kinks, but it should be ready by GTC. You can register your spark on Brev. And so now if youswyx: like remote managed yeah, local hardware. Single pane of glass. Yeah. Yeah. Because Brev can already manage other clouds anyway, right?Vibhu: Yeah, yeah. And you use the spark on Brev as well, right?Nader: Yeah. But yeah, exactly. So, so you, you, so you, you set it up at home you can run the command on it, and then it gets it's essentially it'll appear in your Brev account, and then you can take your laptop to a Starbucks or to a cafe, and you'll continue to use your, you can continue use your spark just like any other cloud node on Brev.Yeah. Yeah. And it's just like a pre-provisioned centerswyx: in yourNader: home. Yeah, exactly.swyx: Yeah. Yeah.Vibhu: Tiny little data center.Nader: Tiny little, the size ofVibhu: your phone.SOL Culture and Dynamo Setupswyx: One more thing before we move on to Kyle. Just have so many Jensen stories and I just love, love mining Jensen stories. Uh, my favorite so far is SOL. Uh, what is, yeah, what is S-O-L-S-O-LNader: is actually, i, I think [00:14:00] of all the lessons I've learned, that one's definitely my favorite.Kyle: It'll always stick with you.Nader: Yeah. Yeah. I, you know, in your startup, everything's existential, right? Like we've, we've run out of money. We were like, on the risk of, of losing payroll, we've had to contract our team because we l ran outta money. And so like, um, because of that you're really always forcing yourself to I to like understand the root cause of everything.If you get a date, if you get a timeline, you know exactly why that date or timeline is there. You're, you're pushing every boundary and like, you're not just say, you're not just accepting like a, a no. Just because. And so as you start to introduce more layers, as you start to become a much larger organization, SOL is is essentially like what is the physics, right?The speed of light moves at a certain speed. So if flight's moving some slower, then you know something's in the way. So before trying to like layer reality back in of like, why can't this be delivered at some date? Let's just understand the physics. What is the theoretical limit to like, uh, how fast this can go?And then start to tell me why. ‘cause otherwise people will start telling you why something can't be done. But actually I think any great leader's goal is just to create urgency. Yeah. [00:15:00] There's an infiniteKyle: create compelling events, right?Nader: Yeah.Kyle: Yeah. So l is a term video is used to instigate a compelling event.You say this is done. How do we get there? What is the minimum? As much as necessary, as little as possible thing that it takes for us to get exactly here and. It helps you just break through a bunch of noise.swyx: Yeah.Kyle: Instantly.swyx: One thing I'm unclear about is, can only Jensen use the SOL card? Like, oh, no, no, no.Not everyone get the b******t out because obviously it's Jensen, but like, can someone else be like, no, likeKyle: frontline engineers use it.Nader: Yeah. Every, I think it's not so much about like, get the b******t out. It's like, it's like, give me the root understanding, right? Like, if you tell me something takes three weeks, it like, well, what's the first principles?Yeah, the first principles. It's like, what's the, what? Like why is it three weeks? What is the actual yeah. What's the actual limit of why this is gonna take three weeks? If you're gonna, if you, if let's say you wanted to buy a new computer and someone told you it's gonna be here in five days, what's the SOL?Well, like the SOL is like, I could walk into a Best Buy and pick it up for you. Right? So then anything that's like beyond that is, and is that practical? Is that how we're gonna, you know, let's say give everyone in the [00:16:00] company a laptop, like obviously not. So then like that's the SOL and then it's like, okay, well if we have to get more than 10, suddenly there might be some, right?And so now we can kind of piece the reality back.swyx: So, so this is the. Paul Graham do things that don't scale. Yeah. And this is also the, what people would now call behi agency. Yeah.Kyle: It's actually really interesting because there's a, there's a second hardware angle to SOL that like doesn't come up for all the org sol is used like culturally at aswyx: media for everything.I'm also mining for like, I think that can be annoying sometimes. And like someone keeps going IOO you and you're like, guys, like we have to be stable. We have to, we to f*****g plan. Yeah.Kyle: It's an interesting balance.Nader: Yeah. I encounter that with like, actually just with, with Alec, right? ‘cause we, we have a new conference so we need to launch, we have, we have goals of what we wanna launch by, uh, by the conference and like, yeah.At the end of the day, where isswyx: this GTC?Nader: Um, well this is like, so we, I mean we did it for CES, we did for GT CDC before that we're doing it for GTC San Jose. So I mean, like every, you know, we have a new moment. Um, and we want to launch something. Yeah. And we want to do so at SOL and that does mean that some, there's some level of prioritization that needs [00:17:00] to happen.And so it, it is difficult, right? I think, um, you have to be careful with what you're pushing. You know, stability is important and that should be factored into S-O-L-S-O-L isn't just like, build everything and let it break, you know, that, that's part of the conversation. So as you're laying, layering in all the details, one of them might be, Hey, we could build this, but then it's not gonna be stable for X, y, z reasons.And so that was like, one of our conversations for CES was, you know, hey, like we, we can get this into early access registering your spark with brev. But there are a lot of things that we need to do in order to feel really comfortable from a security perspective, right? There's a lot of networking involved before we deliver that to users.So it's like, okay. Let's get this to a point where we can at least let people experiment with it. We had it in a booth, we had it in Jensen's keynote, and then let's go iron out all the networking kinks. And that's not easy. And so, uh, that can come later. And so that was the way that we layered that back in.Yeah. ButKyle: It's not really about saying like, you don't have to do the, the maintenance or operational work. It's more about saying, you know, it's kind of like [00:18:00] highlights how progress is incremental, right? Like, what is the minimum thing that we can get to. And then there's SOL for like every component after that.But there's the SOL to get you, get you to the, the starting line. And that, that's usually how it's asked. Yeah. On the other side, you know, like SOL came out of like hardware at Nvidia. Right. So SOL is like literally if we ran the accelerator or the GPU with like at basically full speed with like no other constraints, like how FAST would be able to make a program go.swyx: Yeah. Yeah. Right.Kyle: Soswyx: in, in training that like, you know, then you work back to like some percentage of like MFU for example.Kyle: Yeah, that's a, that's a great example. So like, there's an, there's an S-O-L-M-F-U, and then there's like, you know, what's practically achievable.swyx: Cool. Should we move on to sort of, uh, Kyle's side?Uh, Kyle, you're coming more from the data science world. And, uh, I, I mean I always, whenever, whenever I meet someone who's done working in tabular stuff, graph neural networks, time series, these are basically when I go to new reps, I go to ICML, I walk the back halls. There's always like a small group of graph people.Yes. Absolute small group of tabular people. [00:19:00] And like, there's no one there. And like, it's very like, you know what I mean? Like, yeah, no, like it's, it's important interesting work if you care about solving the problems that they solve.Kyle: Yeah.swyx: But everyone else is just LMS all the time.Kyle: Yeah. I mean it's like, it's like the black hole, right?Has the event horizon reached this yet in nerves? Um,swyx: but like, you know, those are, those are transformers too. Yeah. And, and those are also like interesting things. Anyway, uh, I just wanted to spend a little bit of time on, on those, that background before we go into Dynamo, uh, proper.Kyle: Yeah, sure. I took a different path to Nvidia than that, or I joined six years ago, seven, if you count, when I was an intern.So I joined Nvidia, like right outta college. And the first thing I jumped into was not what I'd done in, during internship, which was like, you know, like some stuff for autonomous vehicles, like heavyweight object detection. I jumped into like, you know, something, I'm like, recommenders, this is popular. Andswyx: yeah, he did RexiKyle: as well.Yeah, Rexi. Yeah. I mean that, that was the taboo data at the time, right? You have tables of like, audience qualities and item qualities, and you're trying to figure out like which member of [00:20:00] the audience matches which item or, or more practically which item matches which member of the audience. And at the time, really it was like we were trying to enable.Uh, recommender, which had historically been like a little bit of a CP based workflow into something that like, ran really well in GPUs. And it's since been done. Like there are a bunch of libraries for Axis that run on GPUs. Uh, the common models like Deeplearning recommendation model, which came outta meta and the wide and deep model, which was used or was released by Google were very accelerated by GPUs using, you know, the fast HBM on the chips, especially to do, you know, vector lookups.But it was very interesting at the time and super, super relevant because like we were starting to get like. This explosion of feeds and things that required rec recommenders to just actively be on all the time. And sort of transitioned that a little bit towards graph neural networks when I discovered them because I was like, okay, you can actually use graphical neural networks to represent like, relationships between people, items, concepts, and that, that interested me.So I jumped into that at [00:21:00] Nvidia and, and got really involved for like two-ish years.swyx: Yeah. Uh, and something I learned from Brian Zaro Yeah. Is that you can just kind of choose your own path in Nvidia.Kyle: Oh my God. Yeah.swyx: Which is not a normal big Corp thing. Yeah. Like you, you have a lane, you stay in your lane.Nader: I think probably the reason why I enjoy being in a, a big company, the mission is the boss probably from a startup guy. Yeah. The missionswyx: is the boss.Nader: Yeah. Uh, it feels like a big game of pickup basketball. Like, you know, if you play one, if you wanna play basketball, you just go up to the court and you're like, Hey look, we're gonna play this game and we need three.Yeah. And you just like find your three. That's honestly for every new initiative that's what it feels like. Yeah.Vibhu: It also like shows, right? Like Nvidia. Just releasing state-of-the-art stuff in every domain. Yeah. Like, okay, you expect foundation models with Nemo tron voice just randomly parakeet.Call parakeet just comes out another one, uh, voice. TheKyle: video voice team has always been producing.Vibhu: Yeah. There's always just every other domain of paper that comes out, dataset that comes out. It's like, I mean, it also stems back to what Nvidia has to do, right? You have to make chips years before they're actually produced.Right? So you need to know, you need to really [00:22:00] focus. TheKyle: design process starts likeVibhu: exactlyKyle: three to five years before the chip gets to the market.Vibhu: Yeah. I, I'm curious more about what that's like, right? So like, you have specialist teams. Is it just like, you know, people find an interest, you go in, you go deep on whatever, and that kind of feeds back into, you know, okay, we, we expect predictions.Like the internals at Nvidia must be crazy. Right? You know? Yeah. Yeah. You know, you, you must. Not even without selling to people, you have your own predictions of where things are going. Yeah. And they're very based, very grounded. Right?Kyle: Yeah. It, it, it's really interesting. So there's like two things that I think that Amed does, which are quite interesting.Uh, one is like, we really index into passion. There's a big. Sort of organizational top sound push to like ensure that people are working on the things that they're passionate about. So if someone proposes something that's interesting, many times they can just email someone like way up the chain that they would find this relevant and say like, Hey, can I go work on this?Nader: It's actually like I worked at a, a big company for a couple years before, uh, starting on my startup journey and like, it felt very weird if you were to like email out of chain, if that makes [00:23:00] sense. Yeah. The emails at Nvidia are like mosh pitsswyx: shoot,Nader: and it's just like 60 people, just whatever. And like they're, there's this,swyx: they got messy like, reply all you,Nader: oh, it's in, it's insane.It's insane. They justKyle: help. You know, Maxim,Nader: the context. But, but that's actually like, I've actually, so this is a weird thing where I used to be like, why would we send emails? We have Slack. I am the entire, I'm the exact opposite. I feel so bad for anyone who's like messaging me on Slack ‘cause I'm so unresponsive.swyx: Your emailNader: Maxi, email Maxim. I'm email maxing Now email is a different, email is perfect because man, we can't work together. I'm email is great, right? Because important threads get bumped back up, right? Yeah, yeah. Um, and so Slack doesn't do that. So I just have like this casino going off on the right or on the left and like, I don't know which thread was from where or what, but like the threads get And then also just like the subject, so you can have like working threads.I think what's difficult is like when you're small, if you're just not 40,000 people I think Slack will work fine, but there's, I don't know what the inflection point is. There is gonna be a point where that becomes really messy and you'll actually prefer having email. ‘cause you can have working threads.You can cc more than nine people in a thread.Kyle: You can fork stuff.Nader: You can [00:24:00] fork stuff, which is super nice and just like y Yeah. And so, but that is part of where you can propose a plan. You can also just. Start, honestly, momentum's the only authority, right? So like, if you can just start, start to make a little bit of progress and show someone something, and then they can try it.That's, I think what's been, you know, I think the most effective way to push anything for forward. And that's both at Nvidia and I think just generally.Kyle: Yeah, there's, there's the other concept that like is explored a lot at Nvidia, which is this idea of a zero billion dollar business. Like market creation is a big thing at Nvidia.Like,swyx: oh, you want to go and start a zero billion dollar business?Kyle: Jensen says, we are completely happy investing in zero billion dollar markets. We don't care if this creates revenue. It's important for us to know about this market. We think it will be important in the future. It can be zero billion dollars for a while.I'm probably minging as words here for, but like, you know, like, I'll give an example. NVIDIA's been working on autonomous driving for a a long time,swyx: like an Nvidia car.Kyle: No, they, they'veVibhu: used the Mercedes, right? They're around the HQ and I think it finally just got licensed out. Now they're starting to be used quite a [00:25:00] bit.For 10 years you've been seeing Mercedes with Nvidia logos driving.Kyle: If you're in like the South San Santa Clara, it's, it's actually from South. Yeah. So, um. Zero billion dollar markets are, are a thing like, you know, Jensen,swyx: I mean, okay, look, cars are not a zero billion dollar market. But yeah, that's a bad example.Nader: I think, I think he's, he's messaging, uh, zero today, but, or even like internally, right? Like, like it's like, uh, an org doesn't have to ruthlessly find revenue very quickly to justify their existence. Right. Like a lot of the important research, a lot of the important technology being developed that, that's kind ofKyle: where research, research is very ide ideologically free at Nvidia.Yeah. Like they can pursue things that they wereswyx: Were you research officially?Kyle: I was never in research. Officially. I was always in engineering. Yeah. We in, I'm in an org called Deep Warning Algorithms, which is basically just how do we make things that are relevant to deep warning go fast.swyx: That sounds freaking cool.Vibhu: And I think a lot of that is underappreciated, right? Like time series. This week Google put out time. FF paper. Yeah. A new time series, paper res. Uh, Symantec, ID [00:26:00] started applying Transformers LMS to Yes. Rec system. Yes. And when you think the scale of companies deploying these right. Amazon recommendations, Google web search, it's like, it's huge scale andKyle: Yeah.Vibhu: You want fast?Kyle: Yeah. Yeah. Yeah. Actually it's, it, I, there's a fun moment that brought me like full circle. Like, uh, Amazon Ads recently gave a talk where they talked about using Dynamo for generative recommendation, which was like super, like weirdly cathartic for me. I'm like, oh my God. I've, I've supplanted what I was working on.Like, I, you're using LMS now to do what I was doing five years ago.swyx: Yeah. Amazing. And let's go right into Dynamo. Uh, maybe introduce Yeah, sure. To the top down and Yeah.Kyle: I think at this point a lot of people are familiar with the term of inference. Like funnily enough, like I went from, you know, inference being like a really niche topic to being something that's like discussed on like normal people's Twitter feeds.It's,Nader: it's on billboardsKyle: here now. Yeah. Very, very strange. Driving, driving, seeing just an inference ad on 1 0 1 inference at scale is becoming a lot more important. Uh, we have these moments like, you know, open claw where you have these [00:27:00] agents that take lots and lots of tokens, but produce, incredible results.There are many different aspects of test time scaling so that, you know, you can use more inference to generate a better result than if you were to use like a short amount of inference. There's reasoning, there's quiring, there's, adding agency to the model, allowing it to call tools and use skills.Dyno sort came about at Nvidia. Because myself and a couple others were, were sort of talking about the, these concepts that like, you know, you have inference engines like VLMS, shelan, tenor, TLM and they have like one single copy. They, they, they sort of think about like things as like one single copy, like one replica, right?Why Scale Out WinsKyle: Like one version of the model. But when you're actually serving things at scale, you can't just scale up that replica because you end up with like performance problems. There's a scaling limit to scaling up replicas. So you actually have to scale out to use a, maybe some Kubernetes type terminology.We kind of realized that there was like. A lot of potential optimization that we could do in scaling out and building systems for data [00:28:00] center scale inference. So Dynamo is this data center scale inference engine that sits on top of the frameworks like VLM Shilling and 10 T lm and just makes things go faster because you can leverage the economy of scale.The fact that you have KV cash, which we can define a little bit later, uh, in all these machines that is like unique and you wanna figure out like the ways to maximize your cash hits or you want to employ new techniques in inference like disaggregation, which Dynamo had introduced to the world in, in, in March, not introduced, it was a academic talk, but beforehand.But we are, you know, one of the first frameworks to start, supporting it. And we wanna like, sort of combine all these techniques into sort of a modular framework that allows you to. Accelerate your inference at scale.Nader: By the way, Kyle and I became friends on my first date, Nvidia, and I always loved, ‘cause like he always teaches meswyx: new things.Yeah. By the way, this is why I wanted to put two of you together. I was like, yeah, this is, this is gonna beKyle: good. It's very, it's very different, you know, like we've, we, we've, we've talked to each other a bunch [00:29:00] actually, you asked like, why, why can't we scale up?Nader: Yeah.Scale Up Limits ExplainedNader: model, you said model replicas.Kyle: Yeah. So you, so scale up means assigning moreswyx: heavier?Kyle: Yeah, heavier. Like making things heavier. Yeah, adding more GPUs. Adding more CPUs. Scale out is just like having a barrier saying, I'm gonna duplicate my representation of the model or a representation of this microservice or something, and I'm gonna like, replicate it Many times.Handle, load. And the reason that you can't scale, scale up, uh, past some points is like, you know, there, there, there are sort of hardware bounds and algorithmic bounds on, on that type of scaling. So I'll give you a good example that's like very trivial. Let's say you're on an H 100. The Maxim ENV link domain for H 100, for most Ds H one hundreds is heus, right?So if you scaled up past that, you're gonna have to figure out ways to handle the fact that now for the GPUs to communicate, you have to do it over Infin band, which is still very fast, but is not as fast as ENV link.swyx: Is it like one order of magnitude, like hundreds or,Kyle: it's about an order of magnitude?Yeah. Okay. Um, soswyx: not terrible.Kyle: [00:30:00] Yeah. I, I need to, I need to remember the, the data sheet here, like, I think it's like about 500 gigabytes. Uh, a second unidirectional for ENV link, and about 50 gigabytes a second unidirectional for Infin Band. I, it, it depends on the, the generation.swyx: I just wanna set this up for people who are not familiar with these kinds of like layers and the trash speedVibhu: and all that.Of course.From Laptop to Multi NodeVibhu: Also, maybe even just going like a few steps back before that, like most people are very familiar with. You see a, you know, you can use on your laptop, whatever these steel viol, lm you can just run inference there. All, there's all, you can, youcan run it on thatVibhu: laptop. You can run on laptop.Then you get to, okay, uh, models got pretty big, right? JLM five, they doubled the size, so mm-hmm. Uh, what do you do when you have to go from, okay, I can get 128 gigs of memory. I can run it on a spark. Then you have to go multi GPU. Yeah. Okay. Multi GPU, there's some support there. Now, if I'm a company and I don't have like.I'm not hiring the best researchers for this. Right. But I need to go [00:31:00] multi-node, right? I have a lot of servers. Okay, now there's efficiency problems, right? You can have multiple eight H 100 nodes, but, you know, is that as a, like, how do you do that efficiently?Kyle: Yeah. How do you like represent them? How do you choose how to represent the model?Yeah, exactly right. That's a, that's like a hard question. Everyone asks, how do you size oh, I wanna run GLM five, which just came out new model. There have been like four of them in the past week, by the way, like a bunch of new models.swyx: You know why? Right? Deep seek.Kyle: No comment. Oh. Yeah, but Ggl, LM five, right?We, we have this, new model. It's, it's like a large size, and you have to figure out how to both scale up and scale out, right? Because you have to find the right representation that you care about. Everyone does this differently. Let's be very clear. Everyone figures this out in their own path.Nader: I feel like a lot of AI or ML even is like, is like this. I think people think, you know, I, I was, there was some tweet a few months ago that was like, why hasn't fine tuning as a service taken off? You know, that might be me. It might have been you. Yeah. But people want it to be such an easy recipe to follow.But even like if you look at an ML model and specificKyle: to you Yeah,Nader: yeah.Kyle: And the [00:32:00] model,Nader: the situation, and there's just so much tinkering, right? Like when you see a model that has however many experts in the ME model, it's like, why that many experts? I don't, they, you know, they tried a bunch of things and that one seemed to do better.I think when it comes to how you're serving inference, you know, you have a bunch of decisions to make and there you can always argue that you can take something and make it more optimal. But I think it's this internal calibration and appetite for continued calibration.Vibhu: Yeah. And that doesn't mean like, you know, people aren't taking a shot at this, like tinker from thinking machines, you know?Yeah. RL as a service. Yeah, totally. It's, it also gets even harder when you try to do big model training, right? We're not the best at training Moes, uh, when they're pre-trained. Like we saw this with LAMA three, right? They're trained in such a sparse way that meta knows there's gonna be a bunch of inference done on these, right?They'll open source it, but it's very trained for what meta infrastructure wants, right? They wanna, they wanna inference it a lot. Now the question to basically think about is, okay, say you wanna serve a chat application, a coding copilot, right? You're doing a layer of rl, you're serving a model for X amount of people.Is it a chat model, a coding model? Dynamo, you know, back to that,Kyle: it's [00:33:00] like, yeah, sorry. So you we, we sort of like jumped off of, you know, jumped, uh, on that topic. Everyone has like, their own, own journey.Cost Quality Latency TradeoffsKyle: And I, I like to think of it as defined by like, what is the model you need? What is the accuracy you need?Actually I talked to NA about this earlier. There's three axes you care about. What is the quality that you're able to produce? So like, are you accurate enough or can you complete the task with enough, performance, high enough performance. Yeah, yeah. Uh, there's cost. Can you serve the model or serve your workflow?Because it's not just the model anymore, it's the workflow. It's the multi turn with an agent cheaply enough. And then can you serve it fast enough? And we're seeing all three of these, like, play out, like we saw, we saw new models from OpenAI that you know, are faster. You have like these new fast versions of models.You can change the amount of thinking to change the amount of quality, right? Produce more tokens, but at a higher cost in a, in a higher latency. And really like when you start this journey of like trying to figure out how you wanna host a model, you, you, you think about three things. What is the model I need to serve?How many times do I need to call it? What is the input sequence link was [00:34:00] the, what does the workflow look like on top of it? What is the SLA, what is the latency SLA that I need to achieve? Because there's usually some, this is usually like a constant, you, you know, the SLA that you need to hit and then like you try and find the lowest cost version that hits all of these constraints.Usually, you know, you, you start with those things and you say you, you kind of do like a bit of experimentation across some common configurations. You change the tensor parallel size, which is a form of parallelismVibhu: I take, it goes even deeper first. Gotta think what model.Kyle: Yes, course,ofKyle: course. It's like, it's like a multi-step design process because as you said, you can, you can choose a smaller model and then do more test time scaling and it'll equate the quality of a larger model because you're doing the test time scaling or you're adding a harness or something.So yes, it, it goes way deeper than that. But from the performance perspective, like once you get to the model you need, you need to host, you look at that and you say, Hey. I have this model, I need to serve it at the speed. What is the right configuration for that?Nader: You guys see the recent, uh, there was a paper I just saw like a few days ago that, uh, if you run [00:35:00] the same prompt twice, you're getting like double Just try itagain.Nader: Yeah, exactly.Vibhu: And you get a lot. Yeah. But the, the key thing there is you give the context of the failed try, right? Yeah. So it takes a shot. And this has been like, you know, basic guidance for quite a while. Just try again. ‘cause you know, trying, just try again. Did you try again? All adviceNader: in life.Vibhu: Just, it's a paper from Google, if I'm not mistaken, right?Yeah,Vibhu: yeah. I think it, it's like a seven bas little short paper. Yeah. Yeah. The title's very cute. And it's just like, yeah, just try again. Give it ask context,Kyle: multi-shot. You just like, say like, hey, like, you know, like take, take a little bit more, take a little bit more information, try and fail. Fail.Vibhu: And that basic concept has gone pretty deep.There's like, um, self distillation, rl where you, you do self distillation, you do rl and you have past failure and you know, that gives some signal so people take, try it again. Not strong enough.swyx: Uh, for, for listeners, uh, who listen to here, uh, vivo actually, and I, and we run a second YouTube channel for our paper club where, oh, that's awesome.Vivo just covered this. Yeah. Awesome. Self desolation and all that's, that's why he, to speed [00:36:00] on it.Nader: I'll to check it out.swyx: Yeah. It, it's just a good practice, like everyone needs, like a paper club where like you just read papers together and the social pressure just kind of forces you to just,Nader: we, we,there'sNader: like a big inference.Kyle: ReadingNader: group at a video. I feel so bad every time. I I, he put it on like, on our, he shared it.swyx: One, one ofNader: your guys,swyx: uh, is, is big in that, I forget es han Yeah, yeah,Kyle: es Han's on my team. Actually. Funny. There's a, there's a, there's a employee transfer between us. Han worked for Nater at Brev, and now he, he's on my team.He wasNader: our head of ai. And then, yeah, once we got in, andswyx: because I'm always looking for like, okay, can, can I start at another podcast that only does that thing? Yeah. And, uh, Esan was like, I was trying to like nudge Esan into like, is there something here? I mean, I don't think there's, there's new infant techniques every day.So it's like, it's likeKyle: you would, you would actually be surprised, um, the amount of blog posts you see. And ifswyx: there's a period where it was like, Medusa hydra, what Eagle, like, youKyle: know, now we have new forms of decode, uh, we have new forms of specula, of decoding or new,swyx: what,Kyle: what are youVibhu: excited? And it's exciting when you guys put out something like Tron.‘cause I remember the paper on this Tron three, [00:37:00] uh, the amount of like post train, the on tokens that the GPU rich can just train on. And it, it was a hybrid state space model, right? Yeah.Kyle: It's co-designed for the hardware.Vibhu: Yeah, go design for the hardware. And one of the things was always, you know, the state space models don't scale as well when you do a conversion or whatever the performance.And you guys are like, no, just keep draining. And Nitron shows a lot of that. Yeah.Nader: Also, something cool about Nitron it was released in layers, if you will, very similar to Dynamo. It's, it's, it's essentially it was released as you can, the pre-training, post-training data sets are released. Yeah. The recipes on how to do it are released.The model itself is released. It's full model. You just benefit from us turning on the GPUs. But there are companies like, uh, ServiceNow took the dataset and they trained their own model and we were super excited and like, you know, celebrated that work.ZoomVibhu: different. Zoom is, zoom is CGI, I think, uh, you know, also just to add like a lot of models don't put out based models and if there's that, why is fine tuning not taken off?You know, you can do your own training. Yeah,Kyle: sure.Vibhu: You guys put out based model, I think you put out everything.Nader: I believe I know [00:38:00]swyx: about base. BasicallyVibhu: without baseswyx: basic can be cancelable.Vibhu: Yeah. Base can be cancelable.swyx: Yeah.Vibhu: Safety training.swyx: Did we get a full picture of dymo? I, I don't know if we, what,Nader: what I'd love is you, you mentioned the three axes like break it down of like, you know, what's prefilled decode and like what are the optimizations that we can get with Dynamo?Kyle: Yeah. That, that's, that's, that's a great point. So to summarize on that three axis problem, right, there are three things that determine whether or not something can be done with inference, cost, quality, latency, right? Dynamo is supposed to be there to provide you like the runtime that allows you to pull levers to, you know, mix it up and move around the parade of frontier or the preto surface that determines is this actually possible with inference And AI todayNader: gives you the knobs.Kyle: Yeah, exactly. It gives you the knobs.Disaggregation Prefill vs DecodeKyle: Uh, and one thing that like we, we use a lot in contemporary inference and is, you know, starting to like pick up from, you know, in, in general knowledge is this co concept of disaggregation. So historically. Models would be hosted with a single inference engine. And that inference engine [00:39:00] would ping pong between two phases.There's prefill where you're reading the sequence generating KV cache, which is basically just a set of vectors that represent the sequence. And then using that KV cache to generate new tokens, which is called Decode. And some brilliant researchers across multiple different papers essentially made the realization that if you separate these two phases, you actually gain some benefits.Those benefits are basically a you don't have to worry about step synchronous scheduling. So the way that an inference engine works is you do one step and then you finish it, and then you schedule, you start scheduling the next step there. It's not like fully asynchronous. And the problem with that is you would have, uh, essentially pre-fill and decode are, are actually very different in terms of both their resource requirements and their sometimes their runtime.So you would have like prefill that would like block decode steps because you, you'd still be pre-filing and you couldn't schedule because you know the step has to end. So you remove that scheduling issue and then you also allow you, or you yourself, to like [00:40:00] split the work into two different ki types of pools.So pre-fill typically, and, and this changes as, as model architecture changes. Pre-fill is, right now, compute bound most of the time with the sequence is sufficiently long. It's compute bound. On the decode side because you're doing a full Passover, all the weights and the entire sequence, every time you do a decode step and you're, you don't have the quadratic computation of KV cache, it's usually memory bound because you're retrieving a linear amount of memory and you're doing a linear amount of compute as opposed to prefill where you retrieve a linear amount of memory and then use a quadratic.You know,Nader: it's funny, someone exo Labs did a really cool demo where for the DGX Spark, which has a lot more compute, you can do the pre the compute hungry prefill on a DG X spark and then do the decode on a, on a Mac. Yeah. And soVibhu: that's faster.Nader: Yeah. Yeah.Kyle: So you could, you can do that. You can do machine strat stratification.Nader: Yeah.Kyle: And like with our future generation generations of hardware, we actually announced, like with Reuben, this [00:41:00] new accelerator that is prefilled specific. It's called Reuben, CPX. SoKubernetes Scaling with GroveNader: I have a question when you do the scale out. Yeah. Is scaling out easier with Dynamo? Because when you need a new node, you can dedicate it to either the Prefill or, uh, decode.Kyle: Yeah. So Dynamo actually has like a, a Kubernetes component in it called Grove that allows you to, to do this like crazy scaling specialization. It has like this hot, it's a representation that, I don't wanna go too deep into Kubernetes here, but there was a previous way that you would like launch multi-node work.Uh, it's called Leader Worker Set. It's in the Kubernetes standard, and Leader worker set is great. It served a lot of people super well for a long period of time. But one of the things that it's struggles with is representing a set of cases where you have a multi-node replica that has a pair, right?You know, prefill and decode, or it's not paired, but it has like a second stage that has a ratio that changes over time. And prefill and decode are like two different things as your workload changes, right? The amount of prefill you'll need to do may change. [00:42:00] The amount of decode that you, you'll need to do might change, right?Like, let's say you start getting like insanely long queries, right? That probably means that your prefill scales like harder because you're hitting these, this quadratic scaling growth.swyx: Yeah.And then for listeners, like prefill will be long input. Decode would be long output, for example, right?Kyle: Yeah. So like decode, decode scale. I mean, decode is funny because the amount of tokens that you produce scales with the output length, but the amount of work that you do per step scales with the amount of tokens in the context.swyx: Yes.Kyle: So both scales with the input and the output.swyx: That's true.Kyle: But on the pre-fold view code side, like if.Suddenly, like the amount of work you're doing on the decode side stays about the same or like scales a little bit, and then the prefilled side like jumps up a lot. You actually don't want that ratio to be the same. You want it to change over time. So Dynamo has a set of components that A, tell you how to scale.It tells you how many prefilled workers and decoded workers you, it thinks you should have, and also provides a scheduling API for Kubernetes that allows you to actually represent and affect this scheduling on, on, on your actual [00:43:00] hardware, on your compute infrastructure.Nader: Not gonna lie. I feel a little embarrassed for being proud of my SVG function earlier.swyx: No, itNader: wasreallyKyle: cute. I, Iswyx: likeNader: it's all,swyx: it's all engineering. It's all engineering. Um, that's where I'mKyle: technical.swyx: One thing I'm, I'm kind of just curious about with all with you see at a systems level, everything going on here. Mm-hmm. And we, you know, we're scaling it up in, in multi, in distributed systems.Context Length and Co Designswyx: Um, I think one thing that's like kind of, of the moment right now is people are asking, is there any SOL sort of upper bounds. In terms of like, let's call, just call it context length for one for of a better word, but you can break it down however you like.Nader: Yeah.swyx: I just think like, well, yeah, I mean, like clearly you can engage in hybrid architectures and throw in some state space models in there.All, all you want, but it looks, still looks very attention heavy.Kyle: Yes. Uh, yeah. Long context is attention heavy. I mean, we have these hybrid models, um,swyx: to take and most, most models like cap out at a million contexts and that's it. Yeah. Like for the last two years has been it.Kyle: Yeah. The model hardware context co-design thing that we're seeing these days is actually super [00:44:00] interesting.It's like my, my passion, like my secret side passion. We see models like Kimmy or G-P-T-O-S-S. I'm use these because I, I know specific things about these models. So Kimmy two comes out, right? And it's an interesting model. It's like, like a deep seek style architecture is MLA. It's basically deep seek, scaled like a little bit differently, um, and obviously trained differently as well.But they, they talked about, why they made the design choices for context. Kimmy has more experts, but fewer attention heads, and I believe a slightly smaller attention, uh, like dimension. But I need to remember, I need to check that. Uh, it doesn't matter. But they discussed this actually at length in a blog post on ji, which is like our pu which is like credit puswyx: Yeah.Kyle: Um, in, in China. Chinese red.swyx: Yeah.Kyle: It's, yeah. So it, it's, it's actually an incredible blog post. Uh, like all the mls people in, in, in that, I've seen that on GPU are like very brilliant, but they, they talk about like the creators of Kimi K two [00:45:00] actually like, talked about it on, on, on there in the blog post.And they say, we, we actually did an experiment, right? Attention scales with the number of heads, obviously. Like if you have 64 heads versus 32 heads, you do half the work of attention. You still scale quadratic, but you do half the work. And they made a, a very specific like. Sort of barter in their system, in their architecture, they basically said, Hey, what if we gave it more experts, so we're gonna use more memory capacity.But we keep the amount of activated experts the same. We increase the expert sparsity, so we have fewer experts act. The ratio to of experts activated to number of experts is smaller, and we decrease the number of attention heads.Vibhu: And kind of for context, what the, what we had been seeing was you make models sparser instead.So no one was really touching heads. You're just having, uh,Kyle: well, they, they did, they implicitly made it sparser.Vibhu: Yeah, yeah. For, for Kimmy. They did,Kyle: yes.Vibhu: They also made it sparser. But basically what we were seeing was people were at the level of, okay, there's a sparsity ratio. You want more total parameters, less active, and that's sparsity.[00:46:00]But what you see from papers, like, the labs like moonshot deep seek, they go to the level of, okay, outside of just number of experts, you can also change how many attention heads and less attention layers. More attention. Layers. Layers, yeah. Yes, yes. So, and that's all basically coming back to, just tied together is like hardware model, co-design, which isKyle: hardware model, co model, context, co-design.Vibhu: Yeah.Kyle: Right. Like if you were training a, a model that was like. Really, really short context, uh, or like really is good at super short context tasks. You may like design it in a way such that like you don't care about attention scaling because it hasn't hit that, like the turning point where like the quadratic curve takes over.Nader: How do you consider attention or context as a separate part of the co-design? Like I would imagine hardware or just how I would've thought of it is like hardware model. Co-design would be hardware model context co-designKyle: because the harness and the context that is produced by the harness is a part of the model.Once it's trained in,Vibhu: like even though towards the end you'll do long context, you're not changing architecture through I see. Training. Yeah.Kyle: I mean you can try.swyx: You're saying [00:47:00] everyone's training the harness into the model.Kyle: I would say to some degree, orswyx: there's co-design for harness. I know there's a small amount, but I feel like not everyone has like gone full send on this.Kyle: I think, I think I think it's important to internalize the harness that you think the model will be running. Running into the model.swyx: Yeah. Interesting. Okay. Bash is like the universal harness,Kyle: right? Like I'll, I'll give. An example here, right? I mean, or just like a, like a, it's easy proof, right? If you can train against a harness and you're using that harness for everything, wouldn't you just train with the harness to ensure that you get the best possible quality out of,swyx: Well, the, uh, I, I can provide a counter argument.Yeah, sure. Which is what you wanna provide a generally useful model for other people to plug into their harnesses, right? So if youKyle: Yeah. Harnesses can be open, open source, right?swyx: Yeah. So I mean, that's, that's effectively what's happening with Codex.Kyle: Yeah.swyx: And, but like you may want like a different search tool and then you may have to name it differently or,Nader: I don't know how much people have pushed on this, but can you.Train a model, would it be, have you have people compared training a model for the for the harness versus [00:48:00] like post training forswyx: I think it's the same thing. It's the same thing. It's okay. Just extra post training. INader: see.swyx: And so, I mean, cognition does this course, it does this where you, you just have to like, if your tool is slightly different, um, either force your tool to be like the tool that they train for.Hmm. Or undo their training for their tool and then Oh, that's re retrain. Yeah. It's, it's really annoying and like,Kyle: I would hope that eventually we hit like a certain level of generality with respect to training newswyx: tools. This is not a GI like, it's, this is a really stupid like. Learn my tool b***h.Like, I don't know if, I don't know if I can say that, but like, you know, um, I think what my point kind of is, is that there's, like, I look at slopes of the scaling laws and like, this slope is not working, man. We, we are at a million token con

Ahmad Sadeddin is the founder and CEO of Corgea. Corgea provides the security tools to find, triage, and fix insecure code. Ahmad shares:- Why you don't need to raise much to find PMF - stay lean: you should surprise people with how few people you are.- What is a small amount to raise? And what team size do you need? - Pivoting during YC and how Corgea found their first customers and the signs of Product Market Fit- The journey to Product Market Fit never stops- How Corgea worked towards Product Market FitThis episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.Links:Ahmad Sadeddin https://www.linkedin.com/in/asadeddin/Corgea https://corgea.com/The Fatal Pinch by Paul Graham https://paulgraham.com/pinch.html

Why is iterating hardware so difficult and what would we do if it came time to start a business.In Episode #515 of 'Meanderings', Juan & I discuss: Clayton Christensen's 'The Innovator's Dilemma' book, why incumbents like IBM and Blockbuster struggled with disruptive shifts, how spin-outs can help large firms explore new markets, whether today's tech giants (NVIDIA, Amazon, Alphabet) are genuinely pivoting faster than past eras, the trap of single‑thesis bets (e.g., x402 via Coinbase/Circle), the difference between wealth and money via Paul Graham's classic essay, my slow‑ship shift toward building something around livestreaming/value-for-value/OpenClaw-style agents, Juan's practical plan to buy and streamline existing local service businesses and the enduring challenge of measuring value in a world awash with AI-generated content. No boostagrams but we do appreciate the streaming!Stan Link: https://stan.store/meremortalsTimeline:(00:00:00) Intro(00:00:36) The Innovator's Dilemma book(00:05:20) From hardware to software: DiSASSter(00:10:58) CapEx arms race: Nvidia up, Apple lagging(00:15:04) Incumbents can't buy their way out every time(00:19:13) Is AI truly disruptive? Capital, energy, and hype checks(00:24:50) Business cycles repeat: pivots, exits, and getting left behind(00:29:34) Investing today: concentration, tech dominance, and copper(00:34:05) Investing is prediction: outcomes vs decisions(00:38:02) Finding exposure: beware tiny bets inside behemoths(00:41:01) Boostagram Lounge and supporter shout-outs(00:42:04) Micropayments, value, and streaming money(00:45:19) Why Lightning may not fit continuous payments(00:49:53) Two paths: analogue community vs full-tilt AI grind(00:53:41) A niche edge: 'human-made' as a selling point(01:03:31) A creator's plan: livestreaming with OpenClaw automation(01:08:02) Work futures: lifestyle businesses and human uniqueness(01:14:58) Zero-to-one vs sustainment: knowing your role(01:20:04) Juan's near-term play: buy, streamline, and bundle SMBs(01:23:40) Wrap-up and sign-off Connect with Mere Mortals:Website: https://www.meremortalspodcasts.com/Discord: https://discord.gg/jjfq9eGReUTwitter/X: https://twitter.com/meremortalspodsInstagram: https://www.instagram.com/meremortalspodcasts/TikTok: https://www.tiktok.com/@meremortalspodcastsValue 4 Value Support:Boostagram: https://www.meremortalspodcasts.com/supportPaypal: https://www.paypal.com/paypalme/meremortalspodcast

Wednesday February 18th, 2026

Gary Tan is the President and CEO of Y Combinator.YC is the startup accelerator behind companies like Airbnb, Stripe, Coinbase, Reddit, Twitch, and thousands more. According to Garry, they've invested in 20% of all startups worth $5B or more started since 2012.Gary has lived every side of the YC ecosystem. He went through YC as a founder, later became a partner, started Initialized Capital where he backed companies like Coinbase and Instacart, and then returned to lead YC.We walk through the different “eras” of YC, from the early Paul Graham and Jessica Livingston days in Cambridge, to scaling in San Francisco, to today's push back toward in person community and what Gary calls “founder mode” for the organization itself.We also talk about why the Bay Area still matters so much for startups, what's happening with California taxes and policy, and why Gary has gotten more involved in local politics to keep it the best place for founders to build companies.Then we go deep on the parts of startups people don't talk about enough. Co-founder conflict, rage quitting, therapy and coaching, and why companies inevitably take on the personality and emotional patterns of their founders.We also cover what YC looks for in applications, how the 13 week batch is structured, how Demo Day really works, how to choose the right investors, and what Gary thinks the next phase of YC looks like, including helping founders even after Series A.At the end, Gary shares his personal AI workflow, including meta prompting, comparing outputs across models, and the tools he uses every day to think and build faster.Try Numeral, the end-to-end platform for sales tax and compliance: https://www.numeral.com⁠Sign-up for Flex Elite with code TURNER, get $1,000: https://form.typeform.com/to/Rx9rTjFzTimestamps:(0:05) Moving from Winnipeg to California as a kid(1:35) How YC interviews work(2:55) The first batch in 2005(6:46) Why YC moved from Boston to SF(8:17) California's Billionaire Tax(11:00) Tech should care about public policies(17:01) Going direct to your audience(20:28) The 2nd Era of YC(24:01) Rage quitting Palantir, learning to understand himself(32:41) Co-founder conflict kills most startups(35:15) Joining YC as a group partner(37:22) Initialized Fund 1 (55x DPI)(39:44) Why Garry went back to lead YC(42:44) YC funds 20% of all $5B+ companies(44:30) Lessons from Brian Chesky(48:01) Garry's thoughts on YC rejection(51:41) How to get into YC(58:03) What it's like inside a 13-week YC batch(1:02:23) 20% of YC is hard tech(1:05:55) YC's 3rd era: founder mode, re-batching(1:07:56) Escaping the matrix(1:11:26) Garry's personal AI stack(1:20:25) Tech optimismReferencedY Combinator: https://www.ycombinator.com/Initialized Capital: https://initialized.com/Torch: https://torch.io/Perplexity: https://www.perplexity.ai/Anthropic: https://www.anthropic.com/OpenAI: https://openai.com/Airbnb: https://www.airbnb.com/Kyle Vogt on his new startup: https://www.youtube.com/watch?v=XQoFbvyWEy8Follow Aaron Levie on X: https://x.com/levieFollow GaryTwitter: https://x.com/garytanLinkedIn: https://www.linkedin.com/in/garytan/Follow TurnerTwitter: https://twitter.com/TurnerNovakLinkedIn: https://www.linkedin.com/in/turnernovakSubscribe to my newsletter to get every episode + the transcript in your inbox every week: https://www.thespl.it/

Not smart or dumb, but novice => competent => proficient => expert => master. Pascal's Wager, formulated by 17th-century philosopher Blaise Pascal, is a pragmatic argument for belief in God, suggesting that wagering on God's existence is the only rational choice. Because the potential reward (heaven/infinite gain) is infinite and the loss (if God does not exist) is finite, it is better to live as if God exists. Plato's "noble lie" (gennaion pseudos) is a foundational myth proposed in The Republic to create social cohesion, unity, and acceptance of hierarchical roles within his ideal state. It convinces citizens that their societal positions—rulers (gold), warriors (silver), or producers (bronze)—are determined by birth, fostering loyalty to the city. Related Blog: https://cloudstreaks.substack.com/p/value-added-intelligence-1-quantity

In this episode, I'm joined by Rebecca Hinds — organizational behavior expert and founder of the Work AI Institute at Glean — for a practical conversation about why meetings deteriorate over time and how to redesign them. Rebecca argues that bad meetings aren't a people problem — they're a systems problem. Without intentional design, meetings default to ego, status signaling, conflict avoidance, and performative participation. Over time, low-value meetings become normalized instead of fixed. Drawing on her research at Stanford University and her leadership of the Work Innovation Lab at Asana, she shares frameworks from her new book, Your Best Meeting Ever, including: The four legitimate purposes of a meeting: decide, discuss, debate, or develop The CEO test for when synchronous time is truly required How to codify shared meeting standards Why leaders must explicitly give permission to leave low-value meetings We also explore leadership, motivation, and the myth that kindness and high standards are opposites. Rebecca explains why effective leaders diagnose what drives each individual — encouragement for some, direct challenge for others — and design environments that support both performance and belonging. Finally, we talk about AI and the future of work. Tools amplify existing culture: strong systems improve, broken systems break faster. Organizations that redesign how work happens — not just what tools they use — will have the advantage. If you want to run better meetings, lead with more clarity, and rethink how collaboration actually happens, this episode is for you. You can find Your Best Meeting Ever at major bookstores and learn more at rebeccahinds.com.  00:00 Start 00:27 Why Meetings Get Worse Over Time Robin references Good Omens and the character Crowley, who designs the M25 freeway to intentionally create frustration and misery. They use this metaphor to illustrate how systems can be designed in ways that amplify dysfunction, whether intentionally or accidentally. The idea is that once dysfunctional systems become normalized, people stop questioning them. They also discuss Cory Doctorow's concept of enshittification, where platforms and systems gradually decline as organizational priorities override user experience. Rebecca connects this pattern directly to meetings, arguing that without intentional design, meetings default to chaos and energy drain. Over time, poorly designed meetings become accepted as inevitable rather than treated as solvable design problems. Rebecca references the Simple Sabotage Field Manual created by the Office of Strategic Services during World War II. The manual advised citizens in occupied territories on how to subtly undermine organizations from within. Many of the suggested tactics involved meetings, including encouraging long speeches, focusing on irrelevant details, and sending decisions to unnecessary committees. The irony is that these sabotage techniques closely resemble common behaviors in modern corporate meetings. Rebecca argues that if meetings were designed from scratch today, without legacy habits and inherited norms, they would likely look radically different. She explains that meetings persist in their dysfunctional form because they amplify deeply human tendencies like ego, status signaling, and conflict avoidance. Rebecca traces her interest in teamwork back to her experience as a competitive swimmer in Toronto. Although swimming appears to be an individual sport, she explains that success is heavily dependent on team structure and shared preparation. Being recruited to swim at Stanford exposed her to an elite, team-first environment that reshaped how she thought about performance. She became fascinated by how a group can become greater than the sum of its parts when the right cultural conditions are present. This experience sparked her long-term curiosity about why organizations struggle to replicate the kind of cohesion often seen in sports. At Stanford, Coach Lee Mauer emphasized that emotional wellbeing and performance were deeply connected. The team included world record holders and Olympians, and the performance standards were extremely high. Despite the intensity, the culture prioritized connection and belonging. Rituals like informal story time around the hot tub helped teammates build relationships beyond performance metrics. Rebecca internalized the lesson that elite performance and strong culture are not opposing forces. She saw firsthand that intensity and warmth can coexist, and that psychological safety can actually reinforce high standards rather than weaken them. Later in her career at Asana, Rebecca encountered the company value of rejecting false trade-offs. This reinforced a lesson she had first learned in swimming, which is that many perceived either-or tensions are not actually unavoidable. She argues that organizations often assume they must choose between performance and happiness, or between kindness and accountability. In her experience, these are false binaries that can be resolved through better design and clearer expectations. She emphasizes that motivated and engaged employees tend to produce higher quality work, making culture a strategic advantage rather than a distraction. Kindness versus ruthlessness in leadership Robin raises the contrast between harsh, fear-based leadership styles and more relational, positive leadership approaches. Both styles have produced winning teams, which raises the question of whether success comes because of the leadership style or despite it. Rebecca argues that resilience and accountability are essential, regardless of tone. She stresses that kindness alone is not sufficient for high performance, but neither is harshness inherently superior. Effective leadership requires understanding what motivates each individual, since some people thrive on encouragement while others crave direct challenge. Rebecca personally identifies with wanting to be pushed and appreciates clarity when her work falls short of expectations. She concludes that the most effective leaders diagnose motivation carefully and design environments that maximize both growth and performance. 08:51 Building the Book-Launch Team: Mentors, Agents, and Choosing the Right Publisher Robin asks Rebecca about the size and structure of the team she assembled to execute the launch successfully. He is especially curious about what the team actually looked like in practice and how coordinated the effort needed to be. He also asks about the meeting cadence and work cadence required to bring a book launch to life at that level. The framing highlights that writing the book is only one phase, while launching it is an entirely different operational challenge. Rebecca explains that the process felt much more organic than it might appear from the outside. She admits that at the beginning, she underestimated the full scope of what a book launch entails. Her original motivation was simple: she believed she had a valuable perspective, wanted to help people, and loved writing. As she progressed deeper into the publishing process, she realized that writing the manuscript was only one piece of a much larger system. The operational and promotional dimensions gradually revealed themselves as a second job layered on top of authorship. Robin emphasizes that writing a book and publishing a book are fundamentally different jobs. Rebecca agrees and acknowledges that the publishing side requires a completely different skill set and infrastructure. The conversation underscores that authorship is creative work, while publishing and launching require strategy, coordination, and business acumen. Rebecca credits her Stanford mentor, Bob Sutton, as a life changing influence throughout the process. He guided her step by step, including decisions around selecting a publisher and choosing an agent. She initially did not plan to work with an agent, but through guidance and reflection, she shifted her perspective. His mentorship helped her ask better questions and approach the process more strategically rather than reactively. Rebecca reflects on an important mindset shift in her career. Earlier in life, she was comfortable being the big fish in a small pond. Over time, she came to believe that she performs better when surrounded by people who are smarter and more experienced than she is. She describes her superpower as working extremely hard and having confidence in that effort. Because of that, she prefers environments where others elevate her thinking and push her further. This philosophy became central to how she built her book launch team. As Rebecca learned more about the moving pieces required for a successful campaign, she became more intentional about who she wanted involved. She sought the best not in terms of prestige alone, but in terms of belief and commitment. She wanted people who would go to bat for her and advocate for the book with genuine enthusiasm. She noticed that some organizations that looked impressive on paper were not necessarily the right fit for her specific campaign. This led her to have extensive conversations with potential editors and publicists before making decisions. Rebecca developed a personal benchmark for evaluating partners. She paid attention to whether they were willing to apply the book's ideas within their own organizations. For her, that signaled authentic belief rather than surface level marketing support. When Simon and Schuster demonstrated early interest in implementing the book's learnings internally, it stood out as meaningful alignment. That commitment suggested they cared about the substance of the work, not just the promotional campaign. As the process unfolded, Rebecca realized that part of her job was learning what questions to ask. Each conversation with potential partners refined her understanding of what she needed. She became more deliberate about building the right bench of people around her. The team was not assembled all at once, but rather shaped through iterative learning and discernment. The launch ultimately reflected both her evolving standards and her commitment to surrounding herself with people who elevated the work. 12:12 Asking Better Questions & Going Asynchronous Robin highlights the tension between the voice of the book and the posture of a first time author entering a major publishing house. He notes that Best Meeting Ever encourages people to assert authority in meetings by asking about agendas, ownership, and structure. At the same time, Rebecca was entering conversations with an established publisher as a new author seeking partnership. The question becomes how to balance clarity and conviction with humility and openness. Robin frames it as showing up with operational authority while still saying you publish books and I want to work with you. Rebecca calls the question insightful and explains that tactically she relied heavily on asking questions. She describes herself as intentionally curious and even nosy because she did not yet know what she did not know. Rather than pretending to have answers, she used inquiry as a way to build authority through understanding. She asked questions asynchronously almost daily, emailing her agent and editor with anything that came to mind. This allowed her to learn the system while also signaling engagement and seriousness. Rebecca explains that most of the heavy lifting happened outside of meetings. By asking questions over email, she clarified information before stepping into synchronous time. Meetings were then reserved for ambiguity, decision making, and issues that required real time collaboration. As a result, the campaign involved very few meetings overall. She had a biweekly meeting with her core team and roughly monthly conversations with her editor. The rest of the coordination happened asynchronously, which aligned with her philosophy about effective meeting design. Rebecca jokes that one hidden benefit of writing a book on meetings is that everyone shows up more prepared and on time. She also felt internal pressure to model the behaviors she was advocating. The campaign therefore became a real world test of her ideas. She emphasizes that she is glad the launch was not meeting heavy and that it reflected the principles in the book. Robin shares a story about their initial connection through David Shackleford. During a short introductory call, he casually offered to spend time discussing book marketing strategies. Rebecca followed up, scheduled time, and took extensive notes during their conversation. After thanking him, she did not continue unnecessary follow up or prolonged discussion. Instead, she quietly implemented many of the practical strategies discussed. Robin later observed bulk sales, bundled speaking engagements, and structured purchase incentives that reflected disciplined execution. Robin emphasizes that generating ideas is relatively easy compared to implementing them. He connects this to Seth Godin's praise that the book is for people willing to do the work. The real difficulty lies not in brainstorming strategies but in consistently executing them. He describes watching Rebecca implement the plan as evidence that she practices what she preaches. Her hard work and disciplined follow through reinforced his confidence in the book before even reading it. Rebecca responds with gratitude and acknowledges that she took his advice seriously. She affirms that several actions she implemented were directly inspired by their conversation. At the same time, the tone remains grounded and collaborative rather than performative. The exchange illustrates her pattern of seeking input, synthesizing it, and then executing independently. Robin transitions toward the theme of self knowledge and its role in leadership and meetings. He connects Rebecca's disciplined execution to her awareness of her own strengths. The earlier theme resurfaces that she sees hard work and follow through as her superpower. The implication is that effective meetings and effective leadership both begin with understanding how you operate best. 17:48 Self-Knowledge at Work Robin shares that he knows he is motivated by carrots rather than sticks. He explains that praise energizes him and improves his performance more than criticism ever could. As a performer and athlete, he appreciates detailed notes and feedback, but encouragement is what unlocks his best work. He contrasts that with experiences like old school ballet training, where harsh discipline did not bring out his strengths. His point is that understanding how you are wired takes experience and reflection. Rebecca agrees that self knowledge is essential and ties it directly to motivation. She argues that the better you understand yourself, the more clearly you can articulate what drives you. Many people, especially early in their careers, do not pause to examine what truly motivates them. She notes that motivation is often intangible and not primarily monetary. For some people it is praise, for others criticism, learning, mastery, collaboration, or autonomy. She also emphasizes that motivation changes over time and shifts depending on organizational context. One of Rebecca's biggest lessons as a manager and contributor is the importance of codifying self knowledge. Writing down what motivates you and how you work best makes it easier to communicate those needs to others. She believes this explicitness is especially critical during times of change. When work is evolving quickly, assumptions about motivation can lead to disengagement. Making preferences visible reduces friction and prevents misalignment. Rebecca references a recent presentation she gave on the dangers of automating the soul of work. She and her mentor Bob Sutton have discussed how organizations risk stripping meaning from roles if they automate without discernment. She points to research showing that many AI startups are automating tasks people would prefer to keep human. The warning is that just because something can be automated does not mean it should be. Without understanding what makes work meaningful for employees, leaders can unintentionally remove the very elements that motivate people. Rebecca believes managers should create explicit user manuals for their team members. These documents outline how individuals prefer to communicate, what motivates them, and what their career aspirations are. She sees this as a practical leadership tool rather than a symbolic exercise. Referring back to these documents helps leaders guide their teams through uncertainty and change. When asked directly, she confirms that she has implemented this practice in previous roles and intends to do so again. When asked about the future of AI, Rebecca avoids making long term predictions. She observes that the most confident forecasters are often those with something to sell. Her shorter term view is that AI amplifies whatever already exists inside an organization. Strong workflows and cultures may improve, while broken systems may become more efficiently broken. She sees organizations over investing in technology while under investing in people and change management. As a result, productivity gains are appearing at the individual level but not consistently at the team or organizational level. Rebecca acknowledges that there is a possible future where AI creates abundance and healthier work life balance. However, she does not believe current evidence strongly supports that outcome in the near term. She does see promising examples of organizations using AI to amplify collaboration and cross functional work. These examples remain rare but signal that a more human centered future is possible. She is cautiously hopeful but not convinced that the most optimistic scenario will unfold automatically. Robin notes that time horizons for prediction have shortened dramatically. Rebecca agrees and says that six months feels like a reasonable forecasting window in the current environment. She observes that the best leaders are setting thresholds for experimentation and failure. Pilots and proofs of concept should fail at a meaningful rate if organizations are truly exploring. Shorter feedback loops allow organizations to learn quickly rather than over commit to fragile long term assumptions. Robin shares a formative story from growing up in his father's small engineering firm, where he was exposed early to office systems and processes. Later, studying in a Quaker community in Costa Rica, he experienced full consensus decision making. He recalls sitting through extended debates, including one about single versus double ply toilet paper. As a fourteen year old who would rather have been climbing trees in the rainforest, the meeting felt painfully misaligned with his energy. That experience contributed to his lifelong desire to make work and collaboration feel less draining and more intentional. The story reinforces the broader theme that poorly designed meetings can disconnect people from purpose and engagement. 28:31 Leadership vs. Tribal Instincts Rebecca explains that much of dysfunctional meeting behavior is rooted in tribal human instincts. People feel loyalty to the group and show up to meetings simply to signal belonging, even when the meeting is not meaningful. This instinct to attend regardless of value reinforces bloated calendars and performative participation. She argues that effective meeting design must actively counteract these deeply human tendencies. Without intentional structure, meetings default to social signaling rather than productive collaboration. Rebecca emphasizes that leadership plays a critical role in changing meeting culture Leaders must explicitly give employees permission to leave meetings when they are not contributing. They must also normalize asynchronous work as a legitimate and often superior alternative. Without that top down permission, employees will continue attending out of fear or habit. Meeting reform requires visible endorsement from those with authority. Power dynamics and pushing back without positional authority Robin reflects on the power of writing a book on meetings while still operating within a hierarchy. He asks how individuals without formal authority can challenge broken systems. Rebecca responds that there is no universal solution because outcomes depend heavily on psychological safety. In organizations with high trust, there is often broad recognition that meetings are ineffective and a desire to fix them. In lower trust environments, change must be approached more strategically and indirectly. Rebecca advises employees to lead with curiosity rather than confrontation. Instead of calling out a bad meeting, one might ask whether their presence is truly necessary. Framing the question around contribution rather than judgment reduces defensiveness. This approach lowers the emotional temperature and keeps the conversation constructive. Curiosity shifts the tone from personal critique to shared problem solving. In psychologically unsafe environments, Rebecca suggests shifting enforcement to systems rather than individuals. Automated rules such as canceling meetings without agendas or without sufficient confirmations can reduce personal friction. When technology enforces standards, it feels less like a personal attack. Codified rules provide employees with shared language and objective criteria. This reduces the perception that opting out is a rejection of the person rather than a rejection of the structure. Rebecca argues that every organization should have a clear and shared definition of what deserves to be a meeting. If five employees are asked what qualifies as a meeting, they should give the same answer. Without explicit criteria, decisions default to habit and hierarchy. Clear rules give employees confidence to push back constructively. Shared standards transform meeting participation from a personal negotiation into a procedural one. Rebecca outlines a two part test to determine whether a meeting should exist. First, the meeting must serve one of four purposes which are to decide, discuss, debate, or develop people. If it does not satisfy one of those four categories, it likely should not be a meeting. Even if it passes that test, it must also satisfy one of the CEO criteria. C refers to complexity and whether the issue contains enough ambiguity to require synchronous dialogue. E refers to emotional intensity and whether reading emotions or managing reactions is important. O refers to one way door decisions, meaning choices that are difficult or costly to reverse. Many organizational decisions are reversible and therefore do not justify synchronous time. Robin asks how small teams without advanced tech stacks can automate meeting discipline. Rebecca explains that many safeguards can be implemented with existing tools such as Google Calendar or simple scripts. Basic rules like requiring an agenda or minimum confirmations can be enforced through standard workflows. Not all solutions require advanced AI tools. The key is introducing friction intentionally to prevent low value meetings from forming. Rebecca notes that more advanced AI tools can measure engagement, multitasking, or participation. Some platforms now provide indicators of attention or involvement during meetings. While these tools are promising, they are not required to implement foundational meeting discipline. She cautions against over investing in shiny tools without first clarifying principles. Metrics are useful when they reinforce intentional design rather than replace it. Rebecca highlights a subtle risk of automation, particularly in scheduling. Tools can be optimized for the sender while increasing friction for recipients. Leaders should consider the system level impact rather than only individual efficiency. Productivity gains at the individual level can create hidden coordination costs for the team. Meeting automation should be evaluated through a collective lens. Rebecca distinguishes between intrusive AI bots that join meetings and simple transcription tools. She is cautious about bots that visibly attend meetings and distract participants. However, she supports consensual transcription when it enhances asynchronous follow up. Effective transcription can reduce cognitive load and free participants to engage more deeply. Used thoughtfully, these tools can strengthen collaboration rather than dilute it. 41:35 Maker vs. Manager: Balancing a Day Job with a Book Launch Robin shares an example from a webinar where attendees were asked for feedback via a short Bitly link before the session closed. He contrasts this with the ineffectiveness of "smiley face/frowny face" buttons in hotel bathrooms—easy to ignore and lacking context. The key is embedding feedback into the process in a way that's natural, timely, and comfortable for participants. Feedback mechanisms should be integrated, low-friction, and provide enough context for meaningful responses. Rebecca recommends a method inspired by Elise Keith called Roti—rating meetings on a zero-to-five scale based on whether they were worth attendees' time. She suggests asking this for roughly 10% of meetings to gather actionable insight. Follow-up question: "What could the organizer do to increase the rating by one point?" This approach removes bias, focuses on attendee experience, and identifies meetings that need restructuring. Splits in ratings reveal misaligned agendas or attendee lists and guide optimization. Robin imagines automating feedback requests via email or tools like Superhuman for convenience. Rebecca agrees and adds that simple forms (Google Forms, paper, or other methods) are effective, especially when anonymous. The goal is simplicity and consistency—given how costly meetings are, there's no excuse to skip feedback. Robin references Paul Graham's essay on maker vs. manager schedules and asks about Rebecca's approach to balancing writing, team coordination, and book marketing. Rebecca shares that 95% of her effort on the book launch was "making"—writing and outreach—thanks to a strong team handling management. She devoted time to writing, scrappy outreach, and building relationships, emphasizing giving without expecting reciprocation. The main coordination challenge was balancing her book work with her full-time job at Asana, requiring careful prioritization. Rebecca created a strict writing schedule inspired by her swimming discipline: early mornings, evenings, and weekends dedicated to writing. She prioritized her book and full-time work while maintaining family commitments. Discipline and clear prioritization were essential to manage competing but synergistic priorities. Robin asks about written vs. spoken communication, referencing Amazon's six-page memos and Zandr Media's phone-friendly quick syncs. Rebecca emphasizes that the answer depends on context but a strong written communication culture is essential in all organizations. Written communication supports clarity, asynchronous work, and complements verbal communication. It's especially important for distributed teams or virtual work. With AI, clear documentation allows better insights, reduces unnecessary content generation, and reinforces disciplined communication. 48:29 AI and the Craft of Writing Rebecca highlights that employees have varying learning preferences—introverted vs. extroverted, verbal vs. written. Effective communication systems should support both verbal and written channels to accommodate these differences. Rebecca's philosophy: writing is a deeply human craft. AI was not used for drafting or creative writing. AI supported research, coordination, tracking trends, and other auxiliary tasks—areas where efficiency is key. Human-led drafting, revising, and word choice remained central to the book. Robin praises Rebecca's use of language, noting it feels human and vivid—something AI cannot replicate in nuance or delight. Rebecca emphasizes that crafting every word, experimenting with phrasing, and tinkering with language is uniquely human. This joy and precision in writing is not replicable by AI and is part of what makes written communication stand out. Rebecca hopes human creativity in writing and oral communication remains valued despite AI advances. Strong written communication is increasingly differentiating for executive communicators and storytellers in organizations. AI can polish or mass-produce text, but human insight, nuance, and storytelling remain essential and career-relevant. Robin emphasizes the importance of reading, writing, and physical activities (like swimming) to reclaim attention from screens. These practices support deep human thinking and creativity, which are harder to replace with AI. Rebecca uses standard tools strategically: email (chunked and batched), Google Docs, Asana, Doodle, and Zoom. Writing is enhanced by switching platforms, fonts, colors, and physical locations—stimulating creativity and perspective. Physical context (plane, café, city) is strongly linked to breakthroughs and memory during writing. Emphasis is on how tools are enacted rather than which tools are used—behavior and discipline matter more than tech. Rebecca primarily recommends business books with personal relevance: Adam Grant's Give and Take – for relational insights beyond work. Bob Sutton's books – for broader lessons on organizational and personal effectiveness. Robert Cialdini's Influence – for understanding human behavior in both professional and personal contexts. Her selections highlight that business literature often offers universal lessons applicable beyond work. 59:48 Where to Find Rebecca The book is available at all major bookstores. Website: rebeccahinds.com LinkedIn: Rebecca Hinds  

When burnout hits, what's the play? We talk how vital a good vacation is with the most qualified person to help you take one. Paul Graham joins the show to talk business, entrepreneurship, and most importantly, that vacation you should have taken last year... and still can.  Connect with Paul for your next trip at https://wishlist.travel.     This episode is sponsored by Prime Payments USA. You've worked hard for your money... so why let another business take what's yours? Go to https://www.primepaymentsusa.com/   Enjoy the show and want to support it? Join our Patreon at Patreon.com/GoodAdvice

We joke about founders wearing many hats, but that metaphor misses the point. It's not about swapping accessories—it's about growing entirely new heads, each with its own brain that thinks, speaks, and prioritizes differently. In this episode, I explore why the transition from consulting or agency work to software entrepreneurship is so disorienting, and why the instincts that made you successful before might be the exact things preventing success now. From the uncomfortable truth about acquisition in low-touch SaaS to the cognitive dissonance of believing in yourself while questioning everything you know, this is about what it really takes to become someone new while staying grounded in who you've always been.This episode of The Bootstraped Founder is sponsored by Paddle.comThe blog post: https://thebootstrappedfounder.com/many-heads-not-many-hats-the-founders-identity-crisis/ The podcast episode: https://tbf.fm/episodes/many-heads-not-many-hats-the-founders-identity-crisisCheck out Podscan, the Podcast database that transcribes every podcast episode out there minutes after it gets released: https://podscan.fmSend me a voicemail on Podline: https://podline.fm/arvidYou'll find my weekly article on my blog: https://thebootstrappedfounder.comPodcast: https://thebootstrappedfounder.com/podcastNewsletter: https://thebootstrappedfounder.com/newsletterMy book Zero to Sold: https://zerotosold.com/My book The Embedded Entrepreneur: https://embeddedentrepreneur.com/My course Find Your Following: https://findyourfollowing.comHere are a few tools I use. Using my affiliate links will support my work at no additional cost to you.- Notion (which I use to organize, write, coordinate, and archive my podcast + newsletter): https://affiliate.notion.so/465mv1536drx- Riverside.fm (that's what I recorded this episode with): https://riverside.fm/?via=arvid- TweetHunter (for speedy scheduling and writing Tweets): http://tweethunter.io/?via=arvid- HypeFury (for massive Twitter analytics and scheduling): https://hypefury.com/?via=arvid60- AudioPen (for taking voice notes and getting amazing summaries): https://audiopen.ai/?aff=PXErZ- Descript (for word-based video editing, subtitles, and clips): https://www.descript.com/?lmref=3cf39Q- ConvertKit (for email lists, newsletters, even finding sponsors): https://convertkit.com?lmref=bN9CZw

SILICON VALLEY KINGMAKER Colleague Keach Hagey, The Optimist. At Stanford, Altman co-founded Loopt, a location-sharing app that won him a meeting with Steve Jobs and a spot in the App Store launch. While Loopt was not a commercial success, the experience taught Altman that his true talent lay in investing and spotting future trends rather than coding. He eventually succeeded Paul Graham as president of Y Combinator, becoming a powerful figure in Silicon Valley who could convince skeptics like Peter Thiel to back his visions. NUMBER 15 SEPTEMBER 1952

SHOW 12-2-2026 THE SHOW BEGIJS WITH DOUBTS ABOUT AI --  a useful invetion that can match the excitement of the first decades of Photography. November 1955 NADAR'S BALLOON AND THE BIRTH OF PHOTOGRAPHY Colleague Anika Burgess, Flashes of Brilliance. In 1863, the photographer Nadar undertook a perilous ascent in a giant balloon to fund experiments for heavier-than-air flight, illustrating the adventurous spirit required of early photographers. This era began with Daguerre's 1839 introduction of the daguerreotype, a process involving highly dangerous chemicals like mercury and iodine to create unique, mirror-like images on copper plates. Pioneers risked their lives using explosive materials to capture reality with unprecedented clarity and permanence. NUMBER 1 PHOTOGRAPHING THE MOON AND SEA Colleague Anika Burgess, Flashes of Brilliance. Early photography expanded scientific understanding, allowing humanity to visualize the inaccessible. James Nasmyth produced realistic images of the moon by photographing plaster models based on telescope observations, aiming to prove its volcanic nature. Simultaneously, Louis Boutan spent a decade perfecting underwater photography, capturing divers in hard-hat helmets. These efforts demonstrated that photography could be a tool for scientific analysis and discovery, revealing details of the natural world previously hidden from the human eye. NUMBER 2 SOCIAL JUSTICE AND NATURE CONSERVATION Colleague Anika Burgess, Flashes of Brilliance. Photography became a powerful agent for social and environmental change. Jacob Riis utilized dangerous flash powder to document the squalid conditions of Manhattan tenements, exposing poverty to the public in How the Other Half Lives. While his methods raised consent issues, they illuminated grim realities. Conversely, Carleton Watkins hauled massive equipment into the wilderness to photograph Yosemite; his majestic images influenced legislation signed by Lincoln to protect the land, proving photography's political impact. NUMBER 3 X-RAYS, SURVEILLANCE, AND MOTION Colleague Anika Burgess, Flashes of Brilliance. The discovery of X-rays in 1895 sparked a "new photography" craze, though the radiation caused severe injuries to early practitioners and subjects. Photography also entered the realm of surveillance; British authorities used hidden cameras to photograph suffragettes, while doctors documented asylum patients without consent. Finally, Eadweard Muybridge's experiments captured horses in motion, settling debates about locomotion and laying the technical groundwork for the future development of motion pictures. NUMBER 4 THE AWAKENING OF CHINA'S ECONOMY Colleague Anne Stevenson-Yang, Wild Ride. Returning to China in 1994, the author witnessed a transformation from the destitute, Maoist uniformity of 1985 to a budding export economy. In the earlier era, workers slept on desks and lacked basic goods, but Deng Xiaoping's realization that the state needed hard currency prompted reforms. Deng established Special Economic Zones like Shenzhen to generate foreign capital while attempting to isolate the population from foreign influence, marking the start of China's export boom. NUMBER 5 RED CAPITALISTS AND SMUGGLERS Colleague Anne Stevenson-Yang, Wild Ride. Following the 1989 Tiananmen crackdown, China reopened to investment in 1992, giving rise to "red capitalists"—often the children of party officials who traded political access for equity. As the central government lost control over local corruption and smuggling rings, it launched "Golden Projects" to digitize and centralize authority over customs and taxes. To avert a banking collapse in 1998, the state created asset management companies to absorb bad loans, effectively rolling over massive debt. NUMBER 6 GHOST CITIES AND THE STIMULUS TRAP Colleague Anne Stevenson-Yang, Wild Ride. China's growth model shifted toward massive infrastructure spending, resulting in "ghost cities" and replica Western towns built to inflate GDP rather than house people. This "Potemkin culture" peaked during the 2008 Olympics, where facades were painted to impress foreigners. To counter the global financial crisis, Beijing flooded the economy with loans, fueling a real estate bubble that consumed more cement in three years than the US did in a century, creating unsustainable debt. NUMBER 7 STAGNATION UNDER SURVEILLANCE Colleague Anne Stevenson-Yang, Wild Ride. The severe lockdowns of the COVID-19 pandemic shattered consumer confidence, leaving citizens insecure and unwilling to spend, which stalled economic recovery. Local governments, cut off from credit and burdened by debt, struggle to provide basic services. Faced with economic stagnation, Xi Jinping has rejected market liberalization in favor of increased surveillance and control, prioritizing regime security over resolving the structural debt crisis or restoring the dynamism of previous decades. NUMBER 8 FAMINE AND FLIGHT TO FREEDOM Colleague Mark Clifford, The Troublemaker. Jimmy Lai was born into a wealthy family that lost everything to the Communist revolution, forcing his father to flee to Hong Kong while his mother endured labor camps. Left behind, Lai survived as a child laborer during a devastating famine where he was perpetually hungry. A chance encounter with a traveler who gave him a chocolate bar inspired him to escape to Hong Kong, the "land of chocolate," stowing away on a boat at age twelve. NUMBER 9 THE FACTORY GUY Colleague Mark Clifford, The Troublemaker. By 1975, Jimmy Lai had risen from a child laborer to a factory owner, purchasing a bankrupt garment facility using stock market profits. Despite being a primary school dropout who learned English from a dictionary, Lai succeeded through relentless work and charm. He capitalized on the boom in American retail sourcing, winning orders from Kmart by producing samples overnight and eventually building Comitex into a leading sweater manufacturer, embodying the Hong Kong dream. NUMBER 10 CONSCIENCE AND CONVERSION Colleague Mark Clifford, The Troublemaker. The 1989 Tiananmen Squaremassacre radicalized Lai, who transitioned from textiles to media, founding Next magazine and Apple Daily to champion democracy. Realizing the brutality of the Chinese Communist Party, he used his wealth to support the student movement and expose regime corruption. As the 1997 handover approached, Lai converted to Catholicism, influenced by his wife and pro-democracy peers, seeking spiritual protection and a moral anchor against the coming political storm. NUMBER 11 PRISON AND LAWFARE Colleague Mark Clifford, The Troublemaker. Following the 2020 National Security Law, authorities raided Apple Daily, froze its assets, and arrested Lai, forcing the newspaper to close. Despite having the means to flee, Lai chose to stay and face imprisonment as a testament to his principles. Now held in solitary confinement, he is subjected to "lawfare"—sham legal proceedings designed to silence him—while he spends his time sketching religious images, remaining a symbol of resistance against Beijing's tyranny. NUMBER 12 FOUNDING OPENAI Colleague Keach Hagey, The Optimist. In 2016, Sam Altman, Greg Brockman, and Ilya Sutskever founded OpenAI as a nonprofit research lab to develop safe artificial general intelligence (AGI). Backed by investors like Elon Musk and Peter Thiel, the organization aimed to be a counterweight to Google's DeepMind, which was driven by profit. The team relied on massive computing power provided by GPUs—originally designed for video games—to train neural networks, recruiting top talent like Sutskever to lead their scientific efforts. NUMBER 13 THE ROOTS OF AMBITION Colleague Keach Hagey, The Optimist. Sam Altman grew up in St. Louis, the son of an idealistic developer and a driven dermatologist mother who instilled ambition and resilience in her children. Altmanattended the progressive John Burroughs School, where his intellect and charisma flourished, allowing him to connect with people on any topic. Though he was a tech enthusiast, his ability to charm others defined him early on, foreshadowing his future as a master persuader in Silicon Valley. NUMBER 14 SILICON VALLEY KINGMAKER Colleague Keach Hagey, The Optimist. At Stanford, Altman co-founded Loopt, a location-sharing app that won him a meeting with Steve Jobs and a spot in the App Store launch. While Loopt was not a commercial success, the experience taught Altman that his true talent lay in investing and spotting future trends rather than coding. He eventually succeeded Paul Graham as president of Y Combinator, becoming a powerful figure in Silicon Valley who could convince skeptics like Peter Thiel to back his visions. NUMBER 15 THE BLIP AND THE FUTURE Colleague Keach Hagey, The Optimist. The viral success of ChatGPT shifted OpenAI's focus from safety to commercialization, despite early internal warnings about the existential risks of AGI. Tensions over safety and Altman's management style led to a "blip" where the nonprofit board fired him, only for him to be quickly reinstated due to employee loyalty. Elon Musk, having lost a power struggle for control of the organization, severed ties, leaving Altman to lead the race toward AGI. NUMBER 16

Listen now: Spotify, Apple and YouTubeIf you've been hearing phrases like “taste is the only thing that will matter for PMs in the AI era” but aren't sure what that actually means—or more importantly, how to build it—this episode is for you.In this conversation, Marc and Ben sit down with Sachin Rekhi, founder, former LinkedIn product leader, and creator of LinkedIn Sales Navigator, to unpack the real mechanics of taste: where it comes from, how to sharpen it, and why it's already the defining skill of AI-native product teams.Sachin shares the frameworks he teaches inside companies and in his Reforge course—from Rick Rubin's “sensitivity & canon” model, to daily design-critique habits, to the patterns he saw across design-driven, metrics-driven, strategy-driven, and sales-driven org cultures.He also tells the untold story of how Sales Navigator went from a tiny skunkworks project to one of LinkedIn's biggest product lines—why social capital mattered, how he managed leadership skepticism, and how he used prototypes, real customer quotes, and narrative-building to secure executive conviction.Whether you're trying to level up your product intuition, navigate organizational taste cultures, or use AI without slipping into “AI slop,” you'll walk away with practical models you can apply immediately to your product work, leadership communication, and team workflows.All episodes of the podcast are also available on Spotify, Apple and YouTube.New to the pod? Subscribe below to get the next episode in your inbox

Liderar uma startup não é o mesmo que gerir uma empresa tradicional. O founder encara o caos, toma decisões sem manual e carrega uma responsabilidade existencial pelo negócio — tudo isso enquanto transforma energia em dinheiro.Neste episódio solo, Pedro Waengertner, CEO da ACE Ventures, explora o que torna a liderança do fundador tão única e por que muitos conselhos de gestão simplesmente não funcionam em contextos empreendedores. A partir de referências como Paul Graham e Brian Chesky (Airbnb), ele compartilha reflexões sobre quando o founder deve se envolver e quando deve, sim, jogar o manual fora.Você vai entender:As diferenças entre fundador, gestor e operadorOs seis perfis de foundersQual combinação desses perfis mais contribui para o crescimento de uma startupPor que autobiografias ensinam mais que playbooksComo usar IA para escalar sua atuação como founderSe você está na linha de frente de um negócio, esse episódio é um lembrete poderoso de que sua visão não é um detalhe — é o que dá vida à empresa.Dá o play e vem com a gente!Materiais mencionados:The Science of Startups: The Impact of Founder Personalities on Company Success — Paul X. McCarthyBefore the Startup — ensaio de Paul Graham (blog post) sobre os desafios de empreender.

What does it take to build tech the world actually trusts? Wikipedia founder Jimmy Wales joins the crew to dig into the real crisis behind AI, social networks, and the web: trust, and how to build it when the stakes are global. Teen founders raise $6M to reinvent pesticides using AI — and convince Paul Graham to join in Introducing SlopStop: Community-driven AI slop detection in Kagi Search Part 1: How I Found Out $1 billion AI company co-founder admits that its $100 a month transcription service was originally 'two guys surviving on pizza' and typing out notes by hand His announcement leaving Meta White House Working on Executive Order to Foil State AI Regulations Nvidia stock soars after results, forecasts top estimates with sales for AI chips 'off the charts' Jeff Bezos Creates A.I. Start-Up Where He Will Be Co-Chief Executive Jack Conte: I'm Building an Algorithm That Doesn't Rot Your Brain AI love, actually Cat island road trip: liquidator's warehouse Gentype The Carpenter's Son... My excerpt from the Q&A Image of the paper Hosts: Leo Laporte, Jeff Jarvis, and Paris Martineau Guest: Jimmy Wales Download or subscribe to Intelligent Machines at https://twit.tv/shows/intelligent-machines. Join Club TWiT for Ad-Free Podcasts! Support what you love and get ad-free shows, a members-only Discord, and behind-the-scenes access. Join today: https://twit.tv/clubtwit Sponsors: ventionteams.com/twit zapier.com/machines agntcy.org spaceship.com/twit

What does it take to build tech the world actually trusts? Wikipedia founder Jimmy Wales joins the crew to dig into the real crisis behind AI, social networks, and the web: trust, and how to build it when the stakes are global. Teen founders raise $6M to reinvent pesticides using AI — and convince Paul Graham to join in Introducing SlopStop: Community-driven AI slop detection in Kagi Search Part 1: How I Found Out $1 billion AI company co-founder admits that its $100 a month transcription service was originally 'two guys surviving on pizza' and typing out notes by hand His announcement leaving Meta White House Working on Executive Order to Foil State AI Regulations Nvidia stock soars after results, forecasts top estimates with sales for AI chips 'off the charts' Jeff Bezos Creates A.I. Start-Up Where He Will Be Co-Chief Executive Jack Conte: I'm Building an Algorithm That Doesn't Rot Your Brain AI love, actually Cat island road trip: liquidator's warehouse Gentype The Carpenter's Son... My excerpt from the Q&A Image of the paper Hosts: Leo Laporte, Jeff Jarvis, and Paris Martineau Guest: Jimmy Wales Download or subscribe to Intelligent Machines at https://twit.tv/shows/intelligent-machines. Join Club TWiT for Ad-Free Podcasts! Support what you love and get ad-free shows, a members-only Discord, and behind-the-scenes access. Join today: https://twit.tv/clubtwit Sponsors: ventionteams.com/twit zapier.com/machines agntcy.org spaceship.com/twit

What does it take to build tech the world actually trusts? Wikipedia founder Jimmy Wales joins the crew to dig into the real crisis behind AI, social networks, and the web: trust, and how to build it when the stakes are global. Teen founders raise $6M to reinvent pesticides using AI — and convince Paul Graham to join in Introducing SlopStop: Community-driven AI slop detection in Kagi Search Part 1: How I Found Out $1 billion AI company co-founder admits that its $100 a month transcription service was originally 'two guys surviving on pizza' and typing out notes by hand His announcement leaving Meta White House Working on Executive Order to Foil State AI Regulations Nvidia stock soars after results, forecasts top estimates with sales for AI chips 'off the charts' Jeff Bezos Creates A.I. Start-Up Where He Will Be Co-Chief Executive Jack Conte: I'm Building an Algorithm That Doesn't Rot Your Brain AI love, actually Cat island road trip: liquidator's warehouse Gentype The Carpenter's Son... My excerpt from the Q&A Image of the paper Hosts: Leo Laporte, Jeff Jarvis, and Paris Martineau Guest: Jimmy Wales Download or subscribe to Intelligent Machines at https://twit.tv/shows/intelligent-machines. Join Club TWiT for Ad-Free Podcasts! Support what you love and get ad-free shows, a members-only Discord, and behind-the-scenes access. Join today: https://twit.tv/clubtwit Sponsors: ventionteams.com/twit zapier.com/machines agntcy.org spaceship.com/twit

What does it take to build tech the world actually trusts? Wikipedia founder Jimmy Wales joins the crew to dig into the real crisis behind AI, social networks, and the web: trust, and how to build it when the stakes are global. Teen founders raise $6M to reinvent pesticides using AI — and convince Paul Graham to join in Introducing SlopStop: Community-driven AI slop detection in Kagi Search Part 1: How I Found Out $1 billion AI company co-founder admits that its $100 a month transcription service was originally 'two guys surviving on pizza' and typing out notes by hand His announcement leaving Meta White House Working on Executive Order to Foil State AI Regulations Nvidia stock soars after results, forecasts top estimates with sales for AI chips 'off the charts' Jeff Bezos Creates A.I. Start-Up Where He Will Be Co-Chief Executive Jack Conte: I'm Building an Algorithm That Doesn't Rot Your Brain AI love, actually Cat island road trip: liquidator's warehouse Gentype The Carpenter's Son... My excerpt from the Q&A Image of the paper Hosts: Leo Laporte, Jeff Jarvis, and Paris Martineau Guest: Jimmy Wales Download or subscribe to Intelligent Machines at https://twit.tv/shows/intelligent-machines. Join Club TWiT for Ad-Free Podcasts! Support what you love and get ad-free shows, a members-only Discord, and behind-the-scenes access. Join today: https://twit.tv/clubtwit Sponsors: ventionteams.com/twit zapier.com/machines agntcy.org spaceship.com/twit

What does it take to build tech the world actually trusts? Wikipedia founder Jimmy Wales joins the crew to dig into the real crisis behind AI, social networks, and the web: trust, and how to build it when the stakes are global. Teen founders raise $6M to reinvent pesticides using AI — and convince Paul Graham to join in Introducing SlopStop: Community-driven AI slop detection in Kagi Search Part 1: How I Found Out $1 billion AI company co-founder admits that its $100 a month transcription service was originally 'two guys surviving on pizza' and typing out notes by hand His announcement leaving Meta White House Working on Executive Order to Foil State AI Regulations Nvidia stock soars after results, forecasts top estimates with sales for AI chips 'off the charts' Jeff Bezos Creates A.I. Start-Up Where He Will Be Co-Chief Executive Jack Conte: I'm Building an Algorithm That Doesn't Rot Your Brain AI love, actually Cat island road trip: liquidator's warehouse Gentype The Carpenter's Son... My excerpt from the Q&A Image of the paper Hosts: Leo Laporte, Jeff Jarvis, and Paris Martineau Guest: Jimmy Wales Download or subscribe to Intelligent Machines at https://twit.tv/shows/intelligent-machines. Join Club TWiT for Ad-Free Podcasts! Support what you love and get ad-free shows, a members-only Discord, and behind-the-scenes access. Join today: https://twit.tv/clubtwit Sponsors: ventionteams.com/twit zapier.com/machines agntcy.org spaceship.com/twit

Most brands focus on what works in November. Smart brands plan for what happens in January. In Part 2 of our Black Friday Growth Series, Jim Huffman shares the strategic lens every DTC operator should adopt before running another BFCM campaign.Following up on the tactical BFCM episode, Jim goes deeper — exploring the downstream effects of your Q4 strategy and how to win long-term. He covers what separates high-ROI brands from revenue-chasers, how to evaluate customer acquisition quality during peak season, and how to balance margin, brand, and lifetime value when everyone else is just trying to “make noise.” This isn't about bigger discounts. It's about smarter growth.TOPICS DISCUSSED IN TODAY'S EPISODEThe biggest mistake brands make during BFCMHow to set Q4 goals that don't backfire in Q1Why who you acquire in Q4 matters more than how manyOffers that build loyalty vs offers that attract deal-chasersHow to use BFCM for email growth and long-term leverageThe mindset shift that separates pro operators from seasonal brandsResources:Growth Marketing OS (Operating System) GrowthHitJim Huffman websiteJim's LinkedinJim's TwitterThe Shopify Growth School Additional episodes you might enjoy:Startup Ideas by Paul Graham (#45)Nathan Barry: How to Bootstrap a Company to $30M in a Crowded Market (#41)How I Met My Biz Partner and Less Learned Hitting $2M ARR (#44)Ryan Hamilton on his Netflix special, touring with Jerry Seinfeld, & how to write a joke (#10)How We're Validating Startup Ideas (#51) 

Black Friday and Cyber Monday aren't just about slapping on a discount. In this episode, Jim Huffman breaks down the offer-led strategies and conversion playbooks top DTC brands use to turn Q4 into their most profitable months - without relying on bloated budgets or ad spend.Originally aired as a guest appearance on the Ecwid eCommerce Show, Jim reveals the full GrowthHit playbook for building winning Black Friday/Cyber Monday campaigns. He dives deep into offer-led growth, conversion rate hacks, retention strategies, email tricks that actually work, and how to survive the Q1 hangover. Whether you're a scrappy DTC founder or scaling a 7-figure Shopify brand, this is your tactical guide to owning Q4 without burning out or discounting your business into the ground.TOPICS DISCUSSED IN TODAY'S EPISODEOffer-led growth: the underrated strategy for converting in Q4The best bundles, BOGOs, and bonus offers that increase AOVTactical email patterns: “Oops” sends, internal leaks, and reminders that convertHow to turn customers into marketers for sustainable growthLanding page and ad setup that avoids performance burnoutWhy Q4 success starts with one hero productReal examples from fashion, consumables, and niche DTC brandsIf you're planning to “wing it” this Black Friday… don't. This episode gives you the framework to build offers, emails, and experiences that drive real growth  in any year. Subscribe for more.Resources:Jim Huffman websiteJim's TwitterGrowthHitThe Growth Marketer's PlaybookThe Shopify Growth School  Additional episodes you might enjoy:Startup Ideas by Paul Graham (#45)Nathan Barry: How to Bootstrap a Company to $30M in a Crowded Market (#41)How I Met My Biz Partner and Less Learned Hitting $2M ARR (#44)Ryan Hamilton on his Netflix special, touring with Jerry Seinfeld, & how to write a joke (#10)How We're Validating Startup Ideas (#51) 

A Note from JamesI'm such a fan of this guy. I loved The Psychology of Money — it felt like he was writing directly about me. I've made a lot of money, lost it all, made it again, lost it again. Over and over. And Morgan gets it.His new book, The Art of Spending Money, hits even deeper. It's not just about being rich; it's about freedom, simplicity, and contentment — the real returns of life. Every word of this conversation is a reminder that money is never about money. It's about independence.Episode DescriptionIn this episode, James sits down with bestselling author Morgan Housel (The Psychology of Money, Same as Ever, The Art of Spending Money) to explore how wealth, happiness, and identity intersect.They talk about why most people spend money to impress strangers who aren't even paying attention, why saving isn't “delayed gratification,” and why independence is the ultimate luxury.Housel and Altucher go beyond finance — into psychology, meaning, and what happens when your identity gets tied up in your success. This is one of the most personal and useful conversations you'll hear about money this year.What You'll LearnWhy the goal of money isn't happiness — it's contentment.How to “purchase independence” instead of possessions.The hidden trap of social signaling and lifestyle inflation.How to build a healthy “psychology of money” that lasts through boom and bust.Why compounding memories might be more valuable than compounding interest.Timestamped Chapters[02:00] “Saving is purchasing independence.”[02:29] Happiness vs. contentment — why wealth brings fewer bad days, not more good ones.[03:00] A Note from James: how Morgan's books mirror his own financial rollercoaster.[04:01] The social trap of spending for admiration.[05:19] Why signaling is universal — and why we overestimate who's watching.[06:29] The three skills of money: making, keeping, and growing it.[07:02] Saving as joy, not sacrifice: how independence is pleasure in the present.[09:08] Why wealth means fewer bad days, not more good ones.[10:00] The quest for the simple life — why simplicity equals freedom.[11:04] James's minimalist experiment: life with one backpack.[12:00] The billionaire's regret — Harvey Firestone and the mansion paradox.[14:15] The psychology of downgrades and why people can't go back.[15:40] Who are you trying to impress? The six people who actually matter.[17:21] Money as a tool vs. money as a scoreboard.[18:35] Why the desire for status falls when you find meaning elsewhere.[21:30] The fear of losing freedom — and how it drives bad decisions.[23:00] Even billionaires worry about losing it all — why fear never goes away.[25:11] Are we wired to worry about money? Nature vs. nurture in financial behavior.[27:39] Envy as outsourced thinking — how jealousy hijacks your decisions.[30:00] The five-minute rule: happiness never lasts, contentment does.[32:00] Saving in your 20s — when it matters and when it doesn't.[33:51] The habits that stick: why early saving teaches independence.[35:29] Why the best memories come when you have the least money.[37:07] Scarcity, gratitude, and why effort creates value.[38:35] Wiping the slate clean: how to escape identity traps.[40:00] Retirement, identity loss, and why former athletes struggle.[42:25] “Keep your identity small.” — lessons from Paul Graham and Tim Ferriss.[45:00] When obsession fuels creation — how James moves between identities.[49:22] Sticking with one thing vs. exploring many — the range paradox.[51:25] The barbell of wisdom: compounding stability vs. compounding experiences.[53:27] The compounding of memories — why they may outlast wealth.[55:15] Simplicity, location, and the emotional geography of memory.Additional Resources

What happens when you build an 8-figure business in one of the most taboo industries - without a team, funding, or traditional ad channels? Brian Sloan did exactly that. And in this episode, he reveals the unfiltered story behind his wild entrepreneurial path. Jim sits down with Brian Sloan, founder of AutoBlow, to unpack how he built a global DTC sex toy brand that now generates 8 figures annually - with a team of just two. From eBay auctions to viral PR stunts, Brian shares how his unconventional path, deep product focus, and scrappy tactics helped him thrive in a space where Facebook ads and mainstream visibility were off-limits. This conversation pulls back the curtain on manufacturing, media manipulation, brand building, and what it really takes to scale when the rules don't apply to your category.Key Topics Covered:How Brian went from selling antiques to latex fetishwear to inventing AutoBlowThe viral crowdfunding stunt that made him internet-famous overnightWhy he ditched Amazon - even after major salesHow to get on GQ, Playboy, Howard Stern, and more without a PR teamThe power of press-worthy product ideasWhy focus (on just 2 SKUs) was his biggest growth unlockBuilding a lean team using a global network of niche freelancersIf you're building in DTC and feel like you're drowning in overhead or noise, this episode is a masterclass in focus, edge, and unconventional growth.Resources:AutoblowJim Huffman websiteJim's TwitterGrowthHitThe Growth Marketer's Playbook Additional episodes you might enjoy:Startup Ideas by Paul Graham (#45)Nathan Barry: How to Bootstrap a Company to $30M in a Crowded Market (#41)How I Met My Biz Partner and Less Learned Hitting $2M ARR (#44)Ryan Hamilton on his Netflix special, touring with Jerry Seinfeld, & how to write a joke (#10)How We're Validating Startup Ideas (#51) 

What do you do when wholesale feels like success but starts killing your brand? For the founders of Mestiza, it meant rewriting the playbook. In this episode, they share how they survived COVID, pivoted to DTC, and built a multi 7-figure fashion business - while raising kids and refusing VC money.Jim is joined by Luisa Takas and Alessandra Perez-Rubio, the powerhouse duo behind Mestiza - a New York-based fashion brand worn by celebrities and loved by loyal customers. The two dive into how they broke into Neiman Marcus early on but quickly realized wholesale wasn't sustainable. From there, they detail how COVID forced a DTC pivot, how their hero product (“The Shimmy Dress”) became a game-changer, and how they've grown profitably while bootstrapping. This is a behind-the-scenes look at resilience, customer obsession, and building a brand with values. TOPICS DISCUSSED IN TODAY'S EPISODEWhy wholesale nearly derailed their brand visionHow COVID forced a DTC rebirth that changed everythingThe power of flagship products like the Shimmy DressTactical tips for customer feedback, crowdfunding, and growthHow they raised a friends-and-family round and used SBA loansWhy being moms made them better foundersThe co-founder dynamic that's lasted longer than most marriagesIf you're building a brand and wondering whether to go DTC, how to grow without VC money, or how to survive the messy middle - this episode is pure gold.Resources:MestizaJim Huffman websiteJim's TwitterGrowthHitThe Growth Marketer's PlaybookAdditional episodes you might enjoy:Startup Ideas by Paul Graham (#45)Nathan Barry: How to Bootstrap a Company to $30M in a Crowded Market (#41)How I Met My Biz Partner and Less Learned Hitting $2M ARR (#44)Ryan Hamilton on his Netflix special, touring with Jerry Seinfeld, & how to write a joke (#10)How We're Validating Startup Ideas (#51) 

In this episode, Robert, co-founder of Wordware, shares the story of turning a sauna brainstorm into one of the largest seed rounds in YC history. He walks us through the early pivots, the highs and lows of YC interviews, and how a viral Twitter Roast fueled their momentum. This is a raw look at conviction, timing, and how founders can compound small wins into historic outcomes. FOUNDER PROFILE: Robert Chandler https://www.linkedin.com/in/robertjhchandler/

Sam Chaudhary is the co-founder and CEO of ClassDojo, a multi-product education platform used in 95% of U.S. schools and over 180 countries globally to connect teachers, students, and families. In this episode, Sam shares the full arc of building ClassDojo, from early skepticism about education and a failed group-making tool, to creating a communication platform loved by millions. In this episode, we discuss: Why ClassDojo was built for consumers (teachers, students and parents) instead of schools How ClassDojo grew entirely by word-of-mouth Sam's unusual approach to building multiple new businesses The founder mindset required to build an industry leader Why relentless resourcefulness is an underrated skill And much more… References: Accel: https://www.accel.com/ Airbnb: https://www.airbnb.com/ Bill Gates: https://www.linkedin.com/in/williamhgates/ Brendan Kereiakes: https://www.linkedin.com/in/product/ ClassDojo: https://www.classdojo.com/ Dominick Bellizzi: https://www.linkedin.com/in/dominickbellizzi/ Geoff Ralston: https://www.linkedin.com/in/geoffralston/ Gonzalo Aguilar Málaga: https://www.linkedin.com/in/gonzalodecheck/ Hamilton Helmer: https://www.linkedin.com/in/hamilton-helmer-42983/ Imagine K12: https://www.imaginek12.com// Khan Academy: https://www.khanacademy.org/ Liam Don: https://www.linkedin.com/in/liamdon/ McKinsey: https://www.mckinsey.com/ Paul Graham: https://x.com/paulg Plaid: https://plaid.com/ Reid Hoffman: https://www.linkedin.com/in/reidhoffman/ Roblox: https://www.roblox.com/ Sal Khan: https://www.linkedin.com/in/khanacademy/ Superhuman: https://superhuman.com/ Tim Brady: https://www.linkedin.com/in/tim-brady-7a632510/ Y Combinator: https://www.ycombinator.com/ Where to find Sam: LinkedIn: https://www.linkedin.com/in/samchaudhary/ Twitter/X: https://x.com/samchaudhary Where to find Brett: LinkedIn: https://www.linkedin.com/in/brett-berson-9986094/ Twitter/X: https://twitter.com/brettberson Timestamps: (01:36) Why education is a “bad market” (02:52) Why enterprise education is broken (03:35) Building for families, not schools (06:53) Early challenges and insights (09:45) Sam's unusual background (11:42) Meeting co-founder Liam at a hackathon (13:22) Getting into Imagine K12 with a group-making tool (19:47) The conversation with Reid Hoffman that changed everything (21:52) Building a network to reach more families (23:30) Scaling by building a community (33:18) Designing for delight and word-of-mouth growth (40:09) Launching the first monetization feature after 7 years (41:35) How to pick markets and when to go broad (46:04) The explosive expansion into the tutoring industry (55:11) Creating safe online spaces for kids (58:01) Harnessing AI in education (59:52) Lessons from ClassDojo's playbook

Ryan Petersen is the founder and CEO of Flexport, the platform that coordinates global logistics from factory floor to customer door. In this conversation, he's refreshingly transparent about the mistakes and painful lessons he's learned building several companies.  He opens up about stepping down as CEO, his struggles with self-confidence, and what happened when he was forced to step in and save his own company.Along the way, we explore why micromanagement might be the secret to better leadership, how Trump-era tariffs reveal the hidden complexity of global trade, and what it takes to scale a company without losing control. There are stories and lessons here you won't find anywhere else, from a data leak that triggered a call from Steve Jobs to flying 500 million masks into the U.S. during a global shutdown.  Thanks to our sponsors for this episode: SHOPIFY: Sign up for your one-dollar-per-month trial period at www.shopify.com/knowledgeproject Basecamp: Stop struggling, start making progress. Get somewhere with Basecamp. Sign up free at www.basecamp.com/knowledgeproject ReMarkable for sponsoring this episode. Get your paper tablet at ⁠reMarkable.com⁠ today Approximate Timestamps: (2:49) Early Life   (4:58) First “Start Up”   (5:38) Living Abroad in China   (10:19) Y Combinator   (11:13) Steve Jobs & the iPhone 3G Launch   (13:41) Lessons from Import Genius   (22:33) Lessons from Paul Graham, Billionaire Investor   (25:31) Flexport Early Days   (36:08) COVID-Era Flexport   (40:06) COVID-Era Flexport – Continued   (44:09) Hiring Flexport's First COO   (47:02) Stepping Down as CEO of Flexport   (51:07) Cutting Cost & Improving Quality   (53:57) Lessons from Other CEOs   (57:05) How to Hire the Best Employees   (59:31) Paul Graham's Closed-Door Talk   (1:03:21) The Value of a 6-Page Monthly Business Review   (1:06:57) Why Do Tariffs Matter?  (1:09:52) Tricks for Dealing with Tariffs   (1:15:43) Other Creative Strategies for Tariffs   (1:21:30) Dealing with Operational Bottlenecks   (1:27:41) Lessons from Charlie Munger (1:30:12) Lessons from Peter Kaufman   (1:37:50) What Is Success for You? Upgrade—If you want to hear my thoughts and reflections at the end of all episodes, join our membership: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠fs.blog/membership⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ and get your own private feed. Newsletter—The Brain Food newsletter delivers actionable insights and thoughtful ideas every Sunday. It takes 5 minutes to read, and it's completely free. Learn more and sign up at ⁠⁠⁠⁠⁠⁠⁠⁠fs.blog/newsletter⁠⁠⁠⁠⁠ Follow me on X at: ⁠⁠⁠⁠⁠x.com/ShaneAParrish Learn more about your ad choices. Visit megaphone.fm/adchoices

Today's show:SYDNEY SWEENEY'S AMERICAN EAGLE AD DIVIDES OUR PANEL!PLUS WHAT STARTUPS CAN LEARN FROM THE VIRAL ASTRONOMER RESPONSEJason, Alex, and Lon are looking at some of the biggest media stories of the day before returning to their favorite topic, tech. Tune in for deep dives on IMAX's new AI film festival, Figma's big IPO and much more!*Timestamps:(0:00) Jason and Alex kick off the show!(3:45) Lon's joining Alex and Lon to discuss the controversial Sydney Sweeney genes/jeans ad(7:20) A look at the polarizing takes on this jeans ad.(10:10) OpenPhone - Streamline and scale your customer communications with OpenPhone. Get 20% off your first 6 months at https://www.openphone.com/twist⁠(11:46) Hear the verdict from Jason, self—proclaimed Chairman of the Interwebs(16:42) Alex points out the economic impact of the controversy(20:26) Vanta - Get $1000 off your SOC 2 at https://www.vanta.com/twist(21:25) Back to the show!(26:35) Paltrow's Astronomer ad, meme-processing and the ideal way to change the conversation(30:04) Vouched - Trust for agents that's built for builders like you. Check it out at http://vouched.id/twist(31:28) Back to the show!(36:48) Netflix and the growing controversy around AI's role in filmmaking(42:53) IMAX/Runway collaboration and Hollywood's shifting attitude toward AI(47:34) Everything that went wrong with the Tea App(54:27) Why Jason thinks app stores should ban “anonymous” forms and message boards(1:04:19) Figma upped its IPO price… what does it mean for the return of liquidity? And is this too high or fairly priced?(1:07:53) Paul Graham says you shouldn't drop out of college to work on a startup… Why Thiel Fellows disagree*Subscribe to the TWiST500 newsletter: https://ticker.thisweekinstartups.comCheck out the TWIST500: https://www.twist500.comSubscribe to This Week in Startups on Apple: https://rb.gy/v19fcp*Follow Lon:X: https://x.com/lons*Follow Alex:X: https://x.com/alexLinkedIn: ⁠https://www.linkedin.com/in/alexwilhelm*Follow Jason:X: https://twitter.com/JasonLinkedIn: https://www.linkedin.com/in/jasoncalacanis*Thank you to our partners:(10:10) OpenPhone - Streamline and scale your customer communications with OpenPhone. Get 20% off your first 6 months at https://www.openphone.com/twist⁠(20:26) Vanta - Get $1000 off your SOC 2 at https://www.vanta.com/twist(30:04) Vouched - Trust for agents that's built for builders like you. Check it out at http://vouched.id/twist*Great TWIST interviews: Will Guidara, Eoghan McCabe, Steve Huffman, Brian Chesky, Bob Moesta, Aaron Levie, Sophia Amoruso, Reid Hoffman, Frank Slootman, Billy McFarland*Check out Jason's suite of newsletters: https://substack.com/@calacanis*Follow TWiST:Twitter: https://twitter.com/TWiStartupsYouTube: https://www.youtube.com/thisweekinInstagram: https://www.instagram.com/thisweekinstartupsTikTok: https://www.tiktok.com/@thisweekinstartupsSubstack: https://twistartups.substack.com*Subscribe to the Founder University Podcast: https://www.youtube.com/@founderuniversity1916