Jean-DenisGreze

LubaHi everyone. Today I'm talking to Jean-Denis Greze, CEO and co-founder of Town, an AI on-ramp platform, and you're going to hear more about what that means. He was the former CTO of Plaid. And Jean-Denis and I talk about so many different things. His decision to be a founder after a very successful C-suite career, the pivot from tax accounting software to an AI assistant, or an AI on-ramp. Jean-Denis is so passionate, enthusiastic, has so many interesting frameworks and stories, and it was a really fun conversation that I hope you enjoy as much as I did. And let's get to it.Link

Amazing. Let's do it. Jean-Denis, thank you so much for taking the time.Link

Jean-DenisThanks for having me. It's so good to see you.Link

LubaLikewise. It's been a long time.Link

Jean-DenisIt's been a long time indeed.Link

LubaWell, there's so much to talk about. First of all, I'm curious. You obviously had a pretty good job with Plaid. You could have gone...Link

Jean-DenisYou're not sure?Link

LubaNo. A great job. A great job at Plaid. You could have gone into C-suite at a similar-stage company or at a bigger company. Why did you decide to experience the stress of being a founder?Link

Jean-DenisYeah, I'm 47 years old, and so when I left Plaid, I was 45, and I felt like if I didn't become a founder now, I don't know if I ever would. You know, I think these jobs are like four-, five-, six-year jobs. And I think if I'd gone and been C-suite somewhere else, I never would have had the energy to start. That was what I was facing, kind of, as a decision.Link

I don't know what you love doing. I love building things for people. And I think the ultimate form of that is to build something from scratch, too. And I'd never done that. I wanted to do it. I felt like I had it in me. I had the energy, the motivation, and, you know, you just got to jump.Link

LubaAnd have you always nurtured the desire to be a founder, or has it become the case in the last few years of your job at Plaid?Link

Jean-DenisYeah, I think... I don't know if I want to say it's a confidence thing, but I felt like I needed to learn for a long time to be able to do it. And I think when I was earlier in my career, I felt a little bit like an outsider. I didn't start my career working in tech in San Francisco. I was an engineer in New York, building software for hedge funds. And even just coming here... You know, I came here to work at Dropbox, starting as an IC at Dropbox and moving to be a manager. My ambition was just, hey, let me work for a great company and try to do a really good job.Link

And then eventually I had a little more ego, and I felt like, hey, maybe I could be a great executive. And so then I started to be like, what kind of company would I like to join and do that? And I joined Plaid, and at Plaid I was incredibly lucky, right? I was there for seven years, but it's one of those companies that just had success after success after success. And then after seven years, I started to feel a little bit tired. I wanted to go back to working on something smaller. And like I said, I was like, hey, if I go smaller, why not just go to the smallest possible thing, right, to nothing, and see if I can get it off the ground?Link

LubaYeah, you mentioned ego. That it was an ego decision.Link

Jean-DenisWell, yeah. No, look, you know what? I think when you try to do these jobs, you have to believe that you can do it. But let me rephrase that. So I think when I joined Plaid, the ego is like, I could be a good VP. I could create a good culture, I could recruit great engineers. That's the ego part. And I think being a founder, the ego is believing that there's something that doesn't exist in the world and you want to create it, you want to make it there. And so the nature of it is, you think if you weren't around, no one else would do it, right? You're bringing something unique to the universe. So for me, that's a little bit of ego, right? I don't think I had that confidence that I could do those kinds of things until I was much later in my career than most people.Link

LubaSo that makes sense. So on this confidence piece, do you feel like a lot of your skills are transferable now that you are a founder? Was it necessary to build up the confidence?Link

Jean-DenisI don't think so. I mean, if someone's listening to this, I'll say the best way to be a founder is to just found, because the role isn't really like any other role. And so you're not going to learn to do it unless you try to do it. I think there's a lot of things from my career that are helpful to be a founder. I have a huge network of engineers, designers, PMs that I've worked with, who respect me and whom I respect, that I can hire. It was probably much easier for me to fundraise than it would be for other people. I do have a ton of experience scaling teams, dealing with people matters. I have a huge network of friends and other founders that can help me. So those are a ton of advantages for sure, but I think they're kind of advantages that increase the odds of success maybe by 5% or by 10%.Link

LubaBut at the end of the day, that's quite a lot.Link

Jean-DenisIt is. But, you know, the reality is, are you cut out to be a founder? That's the hard question. I don't think you find out until you try to do it.Link

LubaHave you found out? Do you think you were cut out to be a founder?Link

Jean-DenisI don't know yet.Link

LubaWhat do you think is going to be that... What are the KPIs here?Link

Jean-DenisI think one KPI... People are always talking, like, why does IPO-ing matter? I think if a business IPOs, it means it lives on beyond any of the founders, right? Because it's owned by shareholders and a board whose responsibility is to find someone to run the company. And so, you know, it's like you've kind of pushed this ship and it can go on its own. So that's one measure of success.Link

I think the one that resonates more with me is how many people are using what we've built and are willing to pay for it, right? If you have ten people use something and pay for it, I don't think you could say you're a very successful founder, right? I think if you have a billion people use something that you've built and are getting value from it, you can say, I'm very, very successful. And there's some number between ten people and a billion that I think I will find fulfilling. I don't know what that number is right now.Link

LubaOkay. So that ambition is not necessarily a billion people, somewhere between a few hundred...Link

Jean-DenisWell, you know, we all like to daydream. And before we go to sleep, we like to have positive thoughts. Certainly I don't think it would be a realistic goal to have right now, given where we are. But I hope we can make a lot of people's lives better, whether we get there or not.Link

LubaYou know, I'm excited to see.Link

Jean-DenisYeah. Me too.Link

LubaDo you wish you started the company earlier?Link

Jean-DenisHmm. This company? I don't know if I could have started this company more than 6 to 12 months earlier, just because of where AI is.Link

LubaYeah, because of where AI is.Link

Jean-DenisBut I think a company... I don't know, I'm not a looking-backwards person. I spend very few cycles wishing I'd done things differently in my life. I've made plenty of mistakes in my personal life and my professional life. So I think if I look back, it would be depressing. I think the question I ask myself is, am I making the right decision now? I've always been very happy. I'm very happy with it.Link

LubaThat's great. I feel like being a happy person is a really big part of having a happy life. And that...Link

Jean-DenisIt's like a choice.Link

LubaYeah, exactly. Yeah. That's true. I hugely believe that every decision we make, even how we look at our own life, is a choice that we personally make. But remember how on Twitter, or on X, the whole thing from Marc Andreessen about introspection went super viral? He was claiming that introspection is not something you should be spending time on. It sounds like that's kind of how you lived your life.Link

Jean-DenisYeah. I think you can definitely ask too many questions, for sure. And I think as a human, sometimes it's easy to find reasons why not, or find criticisms, right? And often you just have to jump and see what happens if you try to do something that maybe you don't have a great plan for or whatnot. So from that perspective, I think that tweet is good. I've never met him, so I don't want to put words... But I think sometimes when he tweets things that might go viral, or that people find interesting, I don't know how true they are...Link

LubaThat was actually from a podcast.Link

Jean-DenisPodcast. So early in my career, until I was in my early 30s, I basically did whatever was easy and in front of me. And I'm reasonably intelligent, and so I had reasonable success. But I think in my early 30s, I started asking myself a little more deliberately, what do I want to do with myself? What skills do I want to develop? What hasn't worked for me before? And I think that took some introspection, and I think I've made better decisions.Link

LubaYeah, better decisions since.Link

Jean-DenisBut I don't self-critique. You know, it's always forward-looking. I'm never like, oh, I should have done something differently before. It's more like, can I learn something that might make me more effective tomorrow?Link

LubaWhat triggered you to ask yourself those questions when you were in your 30s?Link

Jean-DenisWell, I started a career as a lawyer, so...Link

LubaYeah. You have a JD from Harvard, right?Link

Jean-DenisThat's right. So I was an engineer, and then that didn't quite go the way I wanted it to. I basically burnt out as an engineer in grad school, and I was like, yeah, I need to do something different.Link

LubaHow many years were you an engineer at that point?Link

Jean-DenisFor four years.Link

LubaThat's not a long time.Link

Jean-DenisAnd so after grad school, I was like, hey, I don't want to do this. Let me do something the opposite. And the opposite, I think, of engineering sometimes is... It's kind of very structured.Link

LubaYeah. Exactly.Link

Jean-DenisLaw is all gray zone, you know. There's no right or wrong. You're just arguing both sides of an issue. That's an unfair characterization of law, but just go with it. And so then I studied to be a lawyer, and then I worked as a lawyer, and I hated it. Like, literally, I hated it. There's few things in my life that I've not liked.Link

And so I had this one moment, I was like, well, I didn't like my first career, and now I'm not liking my second career. What is it about myself that I enjoy? And that's when I figured out, actually, what I liked about being an engineer is not the programming or the coding. It's talking to customers and building things that solve some problem for them. And that eventually brought me back to becoming an engineer, because it's such a pure version of that. You know, if you're writing software, there's nothing between you and solving the problem, right? You just write the code, and then you can put it in front of somebody and see if it works for them. But I just felt like I'd wasted a little bit of my time, and I needed to be a little more deliberate about it.Link

LubaThat makes a lot of sense. And it worked out.Link

Jean-DenisAnd it worked out, mostly. Still, I feel like it worked out pretty well. Yeah. My life is halfway through. As I said, I'm always happy in the moment.Link

LubaSo obviously Silicon Valley, especially now, glorifies the 18-year-old high school dropout who goes to YC and starts a company. What do you feel like you have as an advantage over that 18-year-old? This is something that you mentioned to me when we were talking about the topics.Link

Jean-DenisWell, what I think I said is, one advantage that I don't have is the ability to work really long hours. Or I think another: I know more about what's not possible, which is, I think, a big disadvantage when you're trying to found something. It's actually kind of good if you don't know what is or isn't possible.Link

LubaYeah.Link

Jean-DenisBecause otherwise, again, you can close off roads just in your head, because you think you know things. When you're younger, you often don't know that much. And so you might explore things that may seem impossible to everyone else but actually lead to something wonderful. So yeah, my advantage is... As I said, could I work 90 hours a week? Sure. But then I would have a divorce and my kids would hate me and, you know, all those things. And I don't want those. I love my wife, I love my kids, I want to spend time with them. And, you know, most 18-year-old founders maybe don't have those things that they can't walk away from in their life.Link

But my advantage is hopefully, and I don't know if these advantages make up for it, but, you know, I can maybe make some slightly better decisions on things that are tried and true. Right? So thinking from first principles, I'm not going to be maybe better at. But, you know, how to handle a hiring process, or how to talk to enterprise customers, or what's the right architecture so that you get privacy and security right, so you don't have a big incident. I've done this stuff for enough decades that I probably have the right priors to make the right decisions along the way. And I think that intuition that's earned through a lot of work has got some value.Link

It might not be valuable for certain types of product, but certainly for a product like Town, where privacy and security are a really key part of the platform, I think it's an advantage. But you know, I don't know, maybe I will lose to an 18-year-old. And if so, well, first, congratulations to the world, because whenever there's competition between different founders trying to do something, you get a better product on the outside. So, you know, I'm fine by losing, if it results in the world having something great.Link

LubaYeah, I think also it's not about losing or winning. You know, there are certain narratives and certain things that are glorified at a given time, but then it's more about the product, the desire, and the ability to create something better for the world. Do you like winning?Link

Jean-DenisI do like winning. We were talking about... I coach these kids. I coach a team of kindergarten soccer players. And we play every weekend, and we don't count the score. And, you know, sportsmanship, sportspersonship, is very important.Link

LubaImportant.Link

Jean-DenisAnd we respect the opponent and all those things. I can tell you, if we win the game, I don't have to count the score. The kids, they count the score, they know, and they have more fun if they're winning. And it's actually something beautiful. Kids that didn't like soccer, and then they experience winning a little bit, or scoring a goal, they want to do more of the thing, and they want to be better at it. So I actually think winning is very important. And I think when I get stressed out, it's not... And winning doesn't mean you're first, right? It just means your product works and a lot of people want it. I stress out a lot about, you know, are we building something that matters? And I think feeling like we're winning is a big part of it.Link

LubaHow do you keep focused on what you guys are doing, versus looking around and keeping the score?Link

Jean-DenisWell, the only score that matters at our stage, early, is: are enough customers willing to use and pay for what we build that we can get to the next, you know, the next launch, maybe the next fundraise, adding a zero to the number of paying customers? That's the only thing that matters.Link

I think when you look around and there's a competitor, and you have reason to believe they are growing faster than you are, then I think for a brief moment of time, it's important to understand what's happening. Is it a different product philosophy? Do they have a different customer ICP? Are they just outmaneuvering you? And I think in those moments, it's not about copying the product. It's understanding what's different about you and them. And sometimes you could be okay with someone moving faster, because you think they're making decisions that for the long term won't be right. Or maybe you realize they have a different customer than the one you're going after, and so it's okay that your paths are diverging.Link

So I don't pay attention to our competitors on a day-to-day basis or a week-to-week basis. But maybe once a month, I like to see, hey, are they going in different directions? Are they going slower or faster? What's happening?Link

LubaYeah. What's happening in the market.Link

Jean-DenisBut day to day, week to week, you can't worry about it, right? You believe your truths about the universe. You build. You put it in front of customers. You see if the truths that you've tried to put in the universe resonate with the user. If it does, great. You think about which things are working. You do more of those, less of the things that aren't. It's kind of that continuous process. And the customer is going to tell you much more what's important to them than you're going to learn from a competitor.Link

LubaYeah, sure.Link

Jean-DenisBecause I talk about lots of things about our startup. And, you know, I'll say what's going well. But there's the secrets about why we're successful. I don't broadcast those. You can't find those out from our blog posts.Link

LubaNow I'm going to ask about that one.Link

Jean-DenisWe're not going to... I'm sorry. We're not going to talk about it on the podcast. Because I think they are... I don't think information really matters that much, but I think there are things we've discovered that are more likely than not to make us successful, that I don't think our competitors see as very important for them. And there's no reason to give that away.Link

LubaTotally. And we're going to spend some time talking about the product, how you guys decided to pivot, so on and so forth. I'm curious, though. What are the things you had to unlearn from your time being a CTO at Plaid, and from just your more corporate career?Link

Jean-DenisUnlearn, unlearn. That implies I learned something. That's a good question.Link

LubaI think from everything you've said, it sounded like you learned quite a bit.Link

Jean-DenisYeah, I don't unlearn. I mean, look, when you're running a 500-person team, your skill set is aligning people, and, you know what I mean? It's like I don't even have to... That stuff is totally irrelevant at our scale. It matters not at all. So I think part of it is, it's not unlearning. It's maybe just amnesia. Just forget all the things that made me successful as an exec. Probably.Link

So one of the differences being a founder versus an exec is the stress. The stress is on a different scale that you wouldn't understand. I wake up, you know, at 4:00 in the morning next to my wife, and I just look, she's sleeping, thank God. I just look at my ceiling, and there's some worry that I have that I just keep worrying about, and worrying about it, worrying about it. And like, why is it that I'm worried...Link

LubaIn a way that you never would have been.Link

Jean-DenisExactly, exactly. I had plenty of stressful things happening to the business, but it was very rare for me to wake up and just be like, I can't sleep. What's happening? So the delta is, if you're an exec and there's a big, hard decision to be made, you can just go talk to your CEO. You can be like, oh, this is a really hard decision. I'm 51/49. But here's what I think we should do. And the responsibility is on someone else.Link

LubaThat's why, as soon as you say the words to them, even if they say, oh, it's your call, you pass on the responsibility.Link

Jean-DenisYeah, I mean, if you're a really good exec, you stand behind it. But, you know, ultimately, if it's a company-ending decision, it's actually never on you. It is on the CEO. That's the nature of the role. So when you're... I have a co-founder. I think for both of us, whatever our titles are, it weighs on us equally. There's really nothing about the company we can't change if we want to. And that's when you realize there's nothing you can't change. And so it's not even about unlearning, but whatever the skills are, if it's going well or not well, you always have this feeling like, what could I do differently that could get it to go better, right?Link

It's little things and big things. If you don't like the food being served at lunch, you don't get to complain, because you could just literally tell people what you want. If someone on the team is a little bit difficult around the edges in how they deal with conflict, well, you could just not hire people like that. You could change the composition of the team however you want to in that moment. If the customers are complaining about the product in a way, you could just be like, that's not our market. We're going after a different market. Literally, the range of possibilities is really unlimited. And so I internalize that as stress. But also, I think, unlike being an exec, because you can pull so many levers, you have to.Link

Yeah. I think you have to flex a much broader view of, you know, how you could get to win. So I'll give you a concrete example. I'm a pretty conservative person in terms of finances and hiring. And so, you know, when you work at a company like Plaid, where the business is very successful but you're trying to get more and more profitable, you have a very tight budget for hiring, and you're like, hey, I can only hire this many people, and this is the ROI of the projects that we have to do. And it's all kind of given to you. Your CFO tells you how this business is supposed to operate.Link

When you're a founder, right now, I could be like, well, do I raise more money and hire 80 people, way, way, way more than our revenue sustains, because this is a market grab and that's how I'm going to win? Or do I do much more conservative, because I think the market's going to go down and I need my unit economics to look really good? Now that's part of the board that I'm playing on. That never would have been the board as an exec. So I don't know. I think that altitude difference is the biggest thing that I've had to adapt to. The skill sets, like recruiting people, motivating people, figuring out what the right technical work is, you know, at a smaller scale, it's different, but it's still some of the same instincts.Link

LubaAs a founder, it's really important to, obviously, one, manage stress, and also figure out what that balance is between what you can actually control and what are the things that are out of your control and out of your hands. And obviously you can say that, okay, more or less everything is in your control. If the market's not working out well, maybe pivot into a different market. But how do you separate the two? What are the techniques, rituals, practices that you do to find that balance for yourself?Link

Jean-DenisI mean, honestly, I think the example you give is a really good one. You could just change market. You control a lot more than you think.Link

LubaThan you think you do.Link

Jean-DenisYeah. There's also... This is a friend of mine who's maybe a little unhinged, but he says that a startup is, you know, the reality distortion field. When you think about it, a startup is: something doesn't exist in the world and you think it should, right? At the very basic of it, there's no thing and you think the thing should exist. And the question is, how do you get many people to believe that they want the thing? That's the job. It's not about anything except convincing other humans that they want the thing. So the whole startup is actually a giant reality distortion field. You're just trying to impose your will on other humans. You're trying to force other humans to want this thing that doesn't exist, right?Link

And so when you think about it that way, I think it's like, you know, if someone doesn't want to buy what you're selling, sorry, you're either not good at selling or you're not selling the right thing. But you fully control those two things. That's kind of the job. On the unit economics side, if it's too expensive to build the thing, you're just not convincing enough people to help you build the thing for cheap enough. At some level, you're just trying to convince humans. There's no law of physics. It's literally just humanity.Link

On the boundary question, I really treat it like anything... I can adapt anything. I think where I carve out exceptions: one, other people on my team don't really like it for me to tell them what to do all day. So I think you always have a balance, which is, how much do you empower other people to make their own mistakes and their own successes on their own? And when do you, as a founder, get involved?Link

The other boundary for me is personal life, family and friends. I think what we do with my wife mostly is, I time-box. So I have these rules: Friday from 6 p.m. until Sunday, 9 a.m., that whole window, not every weekend, but most weekends, I just don't work at all. And if I need to work, I'm very explicit, like, I need four hours on Saturday to work. Because if I know for a time period that I can't be on my phone, I can't be on my laptop, then I can be fully present. We have a date night every Sunday night.Link

LubaThat's amazing.Link

Jean-DenisRight? There's these things, you know, there's the kids' soccer that I coach. There's these things where I'm 100% present. Because otherwise, every... I mean, especially when you're Series A like we are, every incremental minute of effort increases, maybe a very tiny bit, the odds of success. Right. So I feel responsibility towards my team to give them a great outcome and build with them a successful business, right? Because they could go work at Anthropic or OpenAI. The team is that caliber. And so if they're going to spend their time on this company, it's got to be worth it.Link

But also for my family, it's like, I never get the time back with my six-year-old son, right? If there are three days during the week when he needs help with homework, or he wants to play soccer with me outside, or he wants to play video games, or he wants to read a book, if I don't spend that time for two years, then I never saw what he was like as a 6- or 7-year-old. So I don't want that. You just got to be like, hey, what's the right contract between me and my kids and my wife to make those things happen.Link

And the suffering in all of this has not been my family. I mean, I think clearly sometimes they wish I were a little bit more there, but it's mostly actually my friends. I would say my friends get very little of me anymore, because I won't sacrifice family. I won't sacrifice work.Link

LubaSomeone has to go.Link

Jean-DenisYeah, something else has to go. And my health a little bit.Link

LubaSo for someone listening, I know that it's really hard to make the A/B test comparison, but there are a lot of founders who, let's say, are single, don't have any responsibilities, don't have a family. And I'm sure you've seen the meme, "single until Series B"...Link

Jean-DenisMaybe.Link

LubaOkay, there was this big meme where someone was like, oh, I'm staying single until Series B, because every single minute of my day has to go to the company. Would you say that it's net positive to have the support and, you know, this outlet that is your family, your significant other? Or if you were that 25-year-old founder, would you follow the single-until-Series-B advice?Link

Jean-DenisYeah. I mean, more power to any of these trade-offs that someone can make. I don't know if single is what I would say. Kids, pre-kids is a bigger difference than single, married. Well, sorry, in a relationship, married. So single, for me, I don't know, I've not been single for a very long time, so I don't know. But I think when I'm single, I do like to go date, and...Link

LubaYou're still spending that time.Link

Jean-DenisStill spending the time. At the end of the day, I think I feel both a lot of the extremes, like happiness, fun, novel things with my wife, but also a sense of stability that for me personally works really well. But yeah, there's no question I had more time when I was like 22. Right? And I think it's more the discipline of applying the time in the best way possible. Right.Link

Yeah. Also, I feel this about my generation. I was born in 1978. So I think my generation is significantly less mature. When I was 22, I was significantly less mature than a lot of 22-year-old founders that I meet now. And much less disciplined, for sure. I feel like founders that I know that are younger than me, they put much more pressure on themselves to be successful at a young age.Link

LubaBecause we have a lot more examples of people.Link

Jean-DenisYeah, exactly. And also they know so much more about what a startup is supposed to look like. When I graduated college, it was totally foreign that you could become a founder. Bill Gates would have been the example of my generation, and that was alien. Bill Gates was a 1-in-10-year kind of situation, right? And then the one before would have been Steve Jobs.Link

LubaAnd now there's a ton.Link

Jean-DenisYeah, right. And so I think whether it's Michael Truell or Alexandr, you know, from Scale, now a lot of these people are very young, brilliant, mature, well-spoken, good product sense, great engineers. I was nothing like that. So it's a little difficult for me to go back in time. And I'm like, I just couldn't have done it. I was not equipped. But if you are in your 20s and you can stay celibate and just use all your energy on your company, I do think that that does increase the odds of success.Link

I'm actually thinking of a friend of mine who's... Sullivan, who's not celibate. But I think it's like his meme. He had this thing where... what did he say? The power or the water in his apartment ran out, but he didn't find out for like three weeks, because he was sleeping in the office.Link

LubaWow.Link

Jean-DenisYou know, and I was like, that's some commitment. Very good. Power to him.Link

LubaYeah. So good.Link

Jean-DenisThere also, you know, each person is built differently and can do things differently. So, you know, when you and I talk, it's not even in my option space, you know. But that "single until Series B" is very funny. If you look up the meme, they even printed out hats. You can get a hat.Link

LubaThe problem is, you may never get to Series B. Then you just go to the next company.Link

Jean-DenisYeah. You just...Link

LubaOh yeah. I think there's a lot of questions with that commitment. But to each their own. You know, for different people, different things work.Link

Jean-DenisYeah.Link

LubaOkay. Take me to the decision of leaving Plaid to start Town. The first iteration of Town. Did you guys know right away what you were going to be doing? That it's going to be, you know, SMB taxes? Or were you kind of like, YOLO, we'll figure it out?Link

Jean-DenisOh, yeah. So when I left Plaid, I was like 85, 90% sure that I wanted to do a startup.Link

LubaSo it was not 100%.Link

Jean-DenisWell, I mean, I think... You know, your gut is like, what I want to do afterwards. You know, at Plaid, I was so busy thinking about Plaid all the time, I didn't even have an idea. So I was like, I need to decompress. I need to figure out what I really want to do. I think this is what my heart is, but I don't know until I get there. So, you know, I left. I took some time off, a few months off. And then when I came out of that...Link

LubaWhat did you do during your time off?Link

Jean-DenisI went to watch movies at like two in the afternoon. I did some trips with friends. I just did, honestly, whatever I wanted. I drove my kids to school every day and picked them up.Link

I mean, I didn't do anything. You know, I didn't eat, pray, love my way around the world. I didn't buy a Ferrari. I didn't do any of that. But it was just nice to wake up in the morning and not have any plans. And to the extent I had plans, it was like, you know, go play Diablo IV with a friend.Link

LubaYeah, exactly.Link

Jean-DenisIt was fun. It was pure fun, with no expectation that anything would come out of it, you know? But eventually I got very bored, as these things do, because...Link

LubaI think it's important to let yourself get bored to see what sparks.Link

Jean-DenisFor sure. And maybe I should have forced myself to be even more bored. But yeah. So then, I think the 0 to 1 journey, finding the product to work on, is not a fun exercise. You need to be... the batteries, you need to have so much energy, and you really need to think that there's nothing else you want to do in the moment. That was my least favorite part of the startup.Link

LubaOf the process. The 0 to 1.Link

Jean-DenisThat's right. Not talking yourself out of why it's not a good idea. Yeah. And you know, it's silly. You read all these books about blah, blah, blah, and how you talk to them. So yeah, it's just really hard. And it's not science, because if it was science, everybody would do it. Right. It's art. And like all art, you know, it gets pulled out of you very painfully, and sometimes it's garbage.Link

So, but yeah, what happened? So I was going through that process of finding a product to work on. And then actually, a good friend of mine who had built a company in the accounting space basically was like, hey, no one has provided software that helps with taxes for businesses. And there's two different go-to-market angles that we could go after. And so that became one of the 2 to 3 ideas that I validated. And I went to validate it, and it actually looked pretty good.Link

LubaAnd were you validating them with your now co-founder already, or did you guys agree to work together later?Link

Jean-DenisSo the ordering was, this good friend that I knew was like, we should look at the space. And I was like, well, I'm still early. I want to go a little broader. He's like, that's fine. And so I started validating the idea. Then I got more and more excited about it, and then I was like, well, to do this, I think we need different skill sets than the two of us have. And so my current co-founder, Tony, who'd been a good friend, he was also starting to think about his next thing. So we all spent a little bit of time looking at it.Link

And we were like, this is a very concrete problem, which is helping to prepare taxes with AI, but for businesses, right? It's very concrete. There's definitely a pain point. It checked a lot of the boxes. If you use, like, First Round's PMF method, all those things. And so we worked on it for a year, and it mostly worked, in the sense that the technology worked, that we got, after 12 months, pretty high up in the six digits of revenue, and we had customers that liked what we built. But it just didn't feel like it was going to be a huge breakout company.Link

LubaAnd what does it mean, it didn't feel... Tell me more about either the feeling, or how were you guys evaluating?Link

Jean-DenisYeah, selling it was too difficult. Tax is an annual business, meaning you only choose to change who is going to prepare your taxes as a business once a year. And so when you're in a once-a-year business, it's hard to 10x year over year. I mean, it's even hard to 3x year over year, because your sales cycle is in this short window. And even if your users love you and they tell everyone they know, this is the right tax prep solution, no one's going to change their decision until a year later. And we kind of knew that there was always this structural issue, and we looked at a bunch of ways that we could alleviate it, and we just didn't find any that my co-founder and I ended up being really, really optimistic could work.Link

And so what we'd done, actually, when we started the company... It's maybe a little complicated, but there were three founders initially, and then one person left after two months, just because he was burnt out. So there were two of us, Tony and I. We decided, when the third co-founder left, we were like, hey, we have great partners for tax. It's looking really good early on, but one year in, let's put a week in the calendar, and let's evaluate how well this thing is going. And the way I would describe it is, it was going, but it wasn't going to be a rocket ship.Link

LubaNot on fire.Link

Jean-DenisYeah. And we were surrounded by other peers who were also starting companies roughly at the same time that we had, and we saw how they were doing. I've been an angel investor for a long time, and I've seen what happens when you have something that's going okay, and you think you're just a few chess moves from making it great. But sometimes there's nothing you can do with that chess position. You're not in the right part of the board, you know? It's just not the right place to be. We looked at it, and we were like, you know what? We should totally shut it down.Link

LubaBut what do you think? Do you think there is also an internal difference between founders who are like, I'm just going to go balls to the wall on this specific problem? A couple of examples come to mind, like Dylan with Figma, or Notion. You know, they were building for a long time, things didn't work right away, but then they made it work. What is the difference between that type of situation and what you guys were experiencing?Link

Jean-DenisMaybe it was... Well, you don't know, right? Because maybe if we'd stuck with it for another year, it would be crushing it, and I'd be like, acquiring into it in two years. You just don't know. I think for us... It's hard to predict the future. And after the fact, when you look at Figma, you can say, well, it is logical that no one had built software for product designers, and it is logical that no one thought about design as an exercise that's purely about collaboration, so the tool should have collaboration. It's very hard, and you follow those two trends, and if you truly believe in them for a decade, you get this huge outcome. I think for us, the things that you had to believe about the world that would change weren't going to play as much to our advantage.Link

So, for example, I'll give you a good example. We fundamentally believed that all tax prep can be automated by AI. It's true. That was true when we started. We realized how much more true it was after we'd done it for 12 months. So the technology hypothesis was very real. But the problem is, you need the technology hypothesis to also enable a go-to-market hypothesis. And we just didn't have that. With Figma, it's collaborative, so it's going to have network effects. And also, Adobe is not paying attention to product designers, because product design is a tiny percentage of all design, right? With Photoshop, they're looking at brand designers. It's a totally different market. I think you have these trend lines about more and more product being built to help the business.Link

I think the trend lines for us were more like, more automation equals fewer tax preparers. Right. And so you had to believe that AI could allow you to capture the market faster than the market would shrink. And we realized we didn't have in our heads, at that point, a good hypothesis we believed about how we could hyper-grow the business. And it doesn't mean that some other founder wouldn't have it. I think maybe if we'd done a roll-up, it would have worked. There were other shapes that could have gone around that problem. We just stopped getting excited about that problem and the shape of the solutions.Link

And the thing about a startup is, I said before, the world doesn't have a thing. You're trying to force them to want this thing. If you stop believing that you can make the thing happen, it's dead on arrival. You can't do it anymore. And I think in that week, we kind of convinced ourselves that the shape of the opportunity wasn't one that we had clarity on how to solve, and we weren't that excited about it.Link

And, you know, the option space for us is, we still had like $12 million in the bank at a burn of like four. So we had actually unlimited, not unlimited, but quite a lot of runway. And what we knew is that if we committed to another year of tax, it would be very hard to uncommit. Right? Because you're telling these businesses and your partners, hey, we will do tax. You can't stop three months in or six months in. That's rude. That's unethical, right? So we were like, if we're going to do it, we have to do it right now, at the end of tax season. And yeah, that's what we did. We literally pivoted away from a business that was working, because we didn't think we could make it great, just to get back to zero.Link

LubaTake me to that realization and to that moment. What did that week feel like?Link

Jean-DenisThe worst. Yeah. Top three worst weeks of my life. Because you realize it, so then you're like, okay, okay, great. Selfishly, I just don't think this business can work. But what about my seven engineers? What about the five people on the tax team that I've hired? What about the partners? What about our customers? What about admitting to the world, and your friends, and people who respect you or whatever, that you failed? Because that's what had happened. We had failed. And in that moment, you know, you feel like a failure.Link

And then it's like, you know, the stages of grief. You go through them, you try to go through them as quickly as possible, and then you come to a more positive place, which is like, well, actually, none of my engineers would want to work on this if I don't believe in it. They may leave and go somewhere else, but that's better for their life. If I no longer believe in it, I'm not being...Link

LubaWorthy of their respect.Link

Jean-DenisExactly. I'm not being honest. So, you know, you go through the realizations, and then you're like, well, okay, the tax team. There's no tax team if we're not doing tax. So then you're like, okay, well, what if I spend a month helping them find jobs? What if I give them extra long severance because of the work that they've done? What if I let them early exercise, if they believe, and if not, you know, give them the money back from the equity exercise? What can you do to make the situation for everyone be as aboveboard as possible?Link

Of course, at the end of the day, everyone spent three months, six months, 12 months working on something that didn't pan out. But they knew that a little bit, because they decided to join a startup. So, you know, you're like, are we doing right by the people? Yes, we're doing it as right as we can without being bad for shareholder value long term.Link

You realize that, you know, finally, who gives a shit what other people think of you? You know what I mean? It's like, okay, fine. I failed at a startup. No big deal. I did great things before. I will do great things again. It's just life. If you don't take a risk, you don't get anywhere. And so, you know, it took one week to make the decision, and probably another week and a half for us emotionally to get to a place where, let's go. And then as soon as you're at let's go, it's let's go, let's go, let's go. Let's do it fast. Let's go do it well. Treat people with respect, you know, be open and direct about it, and move on.Link

And, you know, people malign investors a lot. I'll tell you, most of the investors were like, thank you, I get it. We've got your back. How can we support you? How can we help you make the move? You know, that's good. I'm lucky that that's kind of the response that we had. And then we did it. And it was sucky for me and for one other person on the team. There was one person on the team, she was incredible, she just helped us wind down every part of the business and all that. So we were in this little bubble of trying to wind down this company.Link

But, you know, everyone else, the week after we made the decision and announced it, they came in, and they were just building product, trying to get to PMF, trying to figure out what was next. A few people decided not to stay on, on the tax team, which is totally fine. But, you know, it was awesome to see people turn the corner, and there you go.Link

LubaSo it sounded like for some people, it was an unexpected decision that you guys decided to turn around.Link

Jean-DenisYeah, it depends how much you read the tea leaves, I would say. Because the thing about it is, I think there was a business to be built. We just would have worked at it for like 5 or 6 years, and it would have been an okay outcome. Yeah. And I think, obviously, because I was doing more of the go-to-market side, I understood the business side. Not everybody on the team does. They're like, well, what are you talking about? The rate of automation that we're doing of AI preparation is going up. We can do more and more of the tax return automatically. It's working.Link

And so I think not everyone kind of stepped back. But honestly, once we talked about it and explained it, you know, 90% of the team was like, okay, let's move forward. I think what was really hard for the team, actually, was the 0 to 1 motion afterwards. And we got lucky, because we got to product-market fit for a new product very quickly.Link

LubaWas the team already with you when you were going from 0 to 1 with the tax product? You were not...Link

Jean-DenisNo. We did 0 to 1...Link

LubaWith you and your co-founder.Link

Jean-DenisMe and my co-founder. And the third. And yeah, and then one person joined really early. But I think most of the rest of the team joined once we knew what we were doing, and we were just building it. So they mostly came to build the thing.Link

And if I had one observation, it's that when you try to recruit people to do tax, and your pitch is big market, hard technical problem, hard to do, but it's bounded, people's mindset is, we're going to execute on this technical thing to solve it. And the team was really good at that. When you go from 0 to 1, it's like, cool, we could build five products. You can build anything. Let's try to get some design partners. And I think not everybody necessarily loves that motion. And that's just life, you know.Link

I will say the team adapted really fast, but I don't think we could have done 12 months of 0 to 1 with a, you know, 6- or 7-person team. I think two co-founders can try to find product-market fit for two years. That's fine. But trying to walk alongside seven other people to do that is tough. And so, again, we got lucky that we got to PMF. But as soon as we got a sniff of PMF, my co-founder, Tony, and I, we were like, you know what? Let's go. This is the thing. Let's go after this thing. Because we knew if we didn't give people a clear goal and timeline, it would just be tough to get everyone motivated. Yeah. I mean, little, little things. Sorry.Link

LubaNo worries. I want to bookmark talking about the actual pivot, and coming into finding product-market fit with Town, what it is right now. I had one last question on this actual feeling of the pivot. For someone listening who is going through a pivot or considering a pivot, how much do you bring your team along to exactly how you're thinking about something? Because it sounds like you guys generally have brought people along, but some people were still surprised, because they weren't necessarily part of the go-to-market motion.Link

Jean-DenisYeah, I think it depends. I think it depends on how far along you are, what your culture is like, a lot of those factors. I think for us, the... That's a good question.Link

So I mean, the best answer that I have for you is, I think in general, it's best to be transparent. And I think, as doubts about the business were shaping up, I wasn't sure if I was being rational or not. Right.Link

LubaWhat do you mean by that?Link

Jean-DenisI didn't know, was I not excited about the business because it wasn't working as well as it could? Or was I not excited about it because I didn't see a way through the maze to get to a big success? I didn't know that, but I started feeling a little bit like something was off. And then we'd always had the time in the calendar to talk about it. And I think any leadership team, you need these moments where you step back, whatever you do. We do it now with the current version. You can't just be in the trenches all day. You're going to miss something. So you need, once a quarter, once a month, once a year, whatever, at some cadence, to step back. So we'd had that.Link

So when Tony and I started to do the walks to do this, and we got to conviction, once we felt like it was the right move, honestly, our reaction was, we need to tell the team ASAP. But it took us another week to get to a place where we're like, okay, let's go. Let's make the change. As soon as we did that, we brought everyone along, literally the next Monday, right? And we hadn't even finished tax season. Meaning, we told everyone, including the tax team. We were like, look, we want you to stick around, and because our clients depend on us, we're going to finish the tax season. It's in another two weeks, right? And this is because we felt like it would have been very dishonest if, for even a 2- to 3-week period, we hadn't told the team.Link

LubaYeah. You're like...Link

Jean-DenisNo, but no one else... It's just, then, I mean, you're not showing up. And I think every day that that happens, then when you do announce the pivot, people are like, well, since when have you known? What the fuck is going on? So I think that's that scenario. I think if it's more of a constant, it's working, it's not working, what's going on, I think there you're better off to be honest, because your team can help get out of the issues. They are creative, right? At startups, you hire people who are creative and smart and who like to solve problems. And so I think if you're clear about the problems and the risks, they may come up with better solutions than you might, right?Link

And I think to that extent, on the team, we had been open about, oh, this deal is taking longer, or this isn't quite working out. I think what they maybe didn't see, and even I didn't fully see until looking at all the data, was the shape of the business overall.Link

LubaMakes sense. Okay, now take me to the actual pivot.Link

Jean-DenisSorry. You know, I just... Pivots are tough, and it's science, right? So, you know, the funny thing is, you just don't know. Maybe we should have pivoted six months earlier. Maybe we shouldn't have done it. You just don't know. Right. And you can only live one version of life. So, you know, you do your best.Link

LubaWise words. Okay, take me to the actual pivot. So you're like, okay, we're pivoting. We can explore whatever we want, or within the five spaces. How does it go, and how do you guys land on an AI assistant?Link

Jean-DenisYeah. So I wouldn't call it an AI assistant, but we could get to that.Link

So when we started, we were like, listen. It was at the time, in the mid-fall of last year, when it was right before Opus 4.5 came out. So it was this time where AI was starting to seem like it was going to jump. And we'd heard rumors about how good the next Opus version was going to be, and how much more agentic approaches, as opposed to more workflow approaches, were going to work. So we told the team, let's spend a few weeks, and, one, just play with the tech, and let's just build a bunch of random prototypes. And then number two, let's ask ourselves if we've built any technology in the last 12 months that would be applicable to other people that are building in this space. And so we did that.Link

And out of that, there's a few more concrete ideas that came about. One of them was a kind of graph database and a vector database together that you could sell kind of as an appliance to a bank or an investment bank, that would basically automatically allow anyone to build agents on top that would find the right information. It was meant to be like, you just connect all your disk drives and everything, and it just magically gives a generic search that's much more effective over your own data.Link

LubaSo it's kind of like a Glean. Like a headless...Link

Jean-DenisYeah. Like a headless Glean, maybe, but with specific application to that space. Another thing that we built, we started playing around with agents around spreadsheets. So not novel ideas, but people were getting excited about just pushing the technology and seeing where it would go. And that was one motion that was just like, hey, let's build some better intuition about where the tech is. Because I do think great ideas, usually there's a go-to-market element and a tech element. And I think we'd been so deep in taxes that we'd forgotten that part of the motion.Link

So we built like 3 or 4 of these products, and then we went to get design partners. And actually one of the design partners was in the hedge fund space. And they wanted to use the library concept, but what they wanted it to do is hook it up to incoming emails. And we didn't build that, but it kind of stayed with me, in my head. And then one weekend I tried to build it, and it was weird. Tony and I were like, it's kind of hard to build an agent over email. And we were like, maybe this also could be a direction that we should pursue for a little bit. So that was more the bottom-up idea that we were going on.Link

And then the second thing I was doing is, I was talking to a bunch of different verticals that we thought could apply some of our technology, and trying to find out if...Link

LubaLike the tax technology?Link

Jean-DenisNo, the underlying tech that powered tax. So for example, we talked to a bunch of companies, and we were like, oh, AI gateways are going to be a thing, and all the governance around AI gateways needs to be much better. So we were having those two types of conversations.Link

LubaI have a quick question. How much of your exploration was driven by what problems you see in a variety of spaces, versus what type of customer you want to work with? The reason why I ask that is because, in my research, in a few interviews that you've done, you very passionately talked about loving working with SMEs, and that you wouldn't want to work in a B2B SaaS place.Link

Jean-DenisYeah, I know. Yeah. So the dream was to find a bottom-up, broadly applicable horizontal product, for a variety of reasons. That's what Dropbox had been. And it's super fun to build those products. I mean, you like Airbnb. It's just super fun and motivating to build something that lots of people can use. And one of the things about taxes, empathy for SMEs was really, really high.Link

But in this period, honestly, we were shooting in a bunch of different directions, because we were trying to understand a bunch of different markets, where it was more enterprise-like B2B, top to bottom, right. And then we were also looking at more horizontal applications of technology for engineers. And then we were just playing with little things on the side. So we were going pretty wide. And my job was to honestly go talk to a bunch of different customer groups and see if there was a prototype somewhere worth building. It was a pretty disorganized motion, I think, as we were coming out of it in December. So after doing it for two months, I think we...Link

LubaWas it just you and your co-founder doing it, or people on the team?Link

Jean-DenisEveryone on the team. Yeah, everyone on the team, for sure. Like, one of the products, for hedge funds, it was literally, after we found a couple of design partners, it was just two engineers working with them to see if we could make...Link

LubaSo the team kind of got broken down.Link

Jean-DenisWe broke teams. Because with seven people, you can't do it with seven. You have to have like 2 to 3 pods. I think it's the only way. And you know, when we were talking to our VCs about it, they were like, well, maybe you should just strip down to just a few folks.Link

LubaYeah, yeah.Link

Jean-DenisAnd I was like, look, I totally get it. But if we can find something with this group of people... The team is so good, we just got to keep it together.Link

LubaWere you worried at all that the team might not be the right team for the eventual product?Link

Jean-DenisNah, maybe. I think, you know, you were talking about advantages earlier. Everyone on the team, basically, the one person in go-to-market, and all the engineers, and the designer, everyone's staff-level. And I mean that not as a weird hierarchy. It's just, these are people who have seen ups and downs. These are people who are good at dealing with a variety of problems. We only have one person on the team that I would say cares about going really, really deep technically. Everyone else is a little bit more of a product engineer mindset. So honestly, it was not a concern.Link

So okay, it sounds super disorganized, and it was. And that's why you should never ask me about how to get to product-market fit. Because what actually happened is, my co-founder and I started playing with this idea of LLMs for email. And in December, we built a prototype. Literally, when an email came in, you could have an agent do something with it. And then we built 3 or 4 core experiences around that that worked well: around meetings, around filtering your email, and drafting replies.Link

And on our website... Our website said, "Town is an applied AI company." We put a sign-in button in the upper right. Didn't say anything else. It was a blue sign-in button. No styling to any of this. You press on the button, it would ask to connect your email, your cal, and your drive to the app. So literally, you would see this screen with nothing.Link

LubaReally?Link

Jean-DenisYeah. And then ten checkboxes. You're like, okay, you connect. And then it would have these basic experiences. And I don't know how, but like 20-ish people decided to click on that and retain. So more people signed up and retained. And so when we came back from the holiday...Link

LubaWhat was the experience then? So you sign up, and then what happens? It drafts similar responses?Link

Jean-DenisIt would just... yeah, it was just draft responses. It would send you a morning briefing, and it would prep you for meetings. That was it. And you could say "new workflow," and then you could type a prompt in English for what you want to happen when an email came in. That was it. That was the product.Link

LubaSo was this around the same time as Poke had launched? Because Poke had a similar experience, but through text.Link

Jean-DenisPoke had launched this summer?Link

LubaOkay.Link

Jean-DenisYeah, Poke had launched, I think, in August or September of the year before. Yeah, this was... We were not paying attention, but this was in the early... OpenClaw was starting a little bit in December to be noticed, or, it was called Clawdbot at the time, in some circles.Link

So I mean, basically, we came back from vacation. We're like, why do we have 20 DAUs? This makes no sense. And we looked at the usage, and there was one person who just built insane things.Link

LubaWow.Link

Jean-DenisNow, you know, we don't look at customer data, but at the time we just saw this person is running like 200 workflows a day. Like, seriously. So we were like, what's going on? And so Tony and I looked at it a little bit, and we were like, you know what? Let's just ship this. So the team came in from vacation, and we were like, we're going to ship this to the world in a month, in February. And the team was like, what's this? We're like, it's the side project that Tony and I have been working on.Link

LubaWow.Link

Jean-DenisAnd it used some of the insights from other things, but it was that. And so the team, God bless their hearts, they came in, and they were like, okay, fine. And they took garbage that had 20 DAUs, and they built it into something that worked. And then we launched. By which I mean I tweeted. X. Sorry, Elon, I'm old. I can't adapt. So we posted it on X, and then we had 100 users, then it went back down to 30. Then a week later it was like 45. Then it was 90. You know what I mean? It just happened. And every day it's kept going up 1 or 2%. That's it. That's literally it.Link

Now, we've learned things since then about why it's working, right? People like to know what's coming in their day. People like to be prepared for meetings. There's lots of jobs, like recruiters or sales, where emails are so similar that AI is actually really good at drafting them. We've built this cool system to write in your voice. We built a system that helps us understand the social network of who you work with and what you work on. There's these building blocks that we got right. We give every assistant an email address, so you can forward things to the assistant. They have names and personalities. There's lots of things that we got right as we started to listen to users and see what they wanted. But the initial thing, honestly, it was just email plus AI. That's it. And lots of people have had that idea, right? And we're still not out of the woods, but, you know, it keeps on growing and resonating with people.Link

LubaSo I signed up for Town, and my Townee's name is Jovan. Jovan Fox.Link

Jean-DenisOkay. And Fox is good. I also have a fox. It tends to like animals.Link

LubaYeah, yeah. But there's a couple of things that stood out to me, and this is my hypothesis for maybe why it worked. One, you know, right away it has all this background info on you as soon as you connect it. It just does things without you doing much. You don't really have to prompt it to do things. You open up your email, it already drafts stuff. You just need to check off what you need help with.Link

Because one of my first questions was, you know, you could set up a skill, a briefing skill, on Claude Cowork, or you could do, you know, all these things with a prompt. But what I'm trying to say is that most people don't really want to bother with finding some prompt and running this job and configuring it, and having it learn, because you have to tell it to learn. Is that what you think?Link

Jean-DenisWelcome to the marketing team. Yeah, that's it. Everything you said. People, look at me, I'm wearing sweatpants, right? People are lazy. Yeah, not in a bad way. People just want the path that has the least resistance, like water going down a hill. So the path of least resistance is not create a skill, blah, blah, blah. The path of least resistance is, we give you something that does it pretty well. And then after that, we can show you that if you just say, oh, next time you do this, can you make the invite have a title that looks like that, it learns that from you. And in the background, yeah, it's a skill, but you don't need to know that. You just need to do the thing that you wanted to. So a lot of it has been getting a default experience for people that's better than nothing, right?Link

LubaFair.Link

Jean-DenisNo one's gonna go and set up an OpenClaw. Most people, like, really very smart, knowledgeable people, just aren't gonna do any of that stuff.Link

LubaYeah, 100%. I mean, I am, you know, not like an average... don't want to do any work, but I'm also pretty lazy. I don't really want to be spending a lot of time on these workflows. I have a bunch of projects on Claude, like a second brain project, and lots of content projects. And I'm like, wow, the bottleneck is still my time. It's the same thing as my aspirations to set up a Roam Research, you know, or a complicated Notion database. It all sounds great in theory, but if I'm the bottleneck that has to spend two hours a week recursively asking it to go learn and rate itself and giving it context, it's just a lot of work.Link

Jean-DenisYeah. The pithy way that Tony says it, he says, if AI is so smart, why do I have to learn it? Why can't it learn me? You know, just learn me. You're supposed to be AI. You're supposed to know all this stuff. Can't you figure out what I want? And I think, end game, if we can pull that string for a long time and be better at suggesting and automating things for people without them having to do much work, I think there's a big product to be built. I think it's not a deep insight. More and more products are doing suggestions, but I think there's real know-how to do this well. And so I'm pretty excited about where we can go if we keep doing it that way.Link

So, you know, you said earlier, and I was like, oh, we're not an AI assistant.Link

LubaI literally just wanted to ask you about that. What are you?Link

Jean-DenisWell, the assistant is how you use and do things, but the product is getting it to do things without you having to learn anything. So actually, the way I think about it in my head is, we help normal people onboard onto AI. Our job is to onboard someone onto getting leverage out of AI. That's it. We're an onboarding product, but it's onboarding for your whole life.Link

LubaSo it's almost like an on-ramp. You know how in crypto there's on-ramps?Link

Jean-DenisYes. It's an on-ramp. Yes. Yes, okay. Maybe...Link

LubaSee, I should be... that should be...Link

Jean-DenisYou know, well, I definitely shouldn't be doing marketing.Link

LubaYeah.Link

Jean-DenisI mean, I think, look, what I'm excited about, and this is like the VC talking, is I think actually, as the models get better, if we build the right structure around how we suggest things, we can do more and more helpful things for folks. Right? You start by organizing a meeting, and then helping someone go into a meeting well-prepped, looking at the action items that come out and reminding them to do it. But maybe some of the action items, you just start doing them until the right step, and you're like, hey, I was going to do this for things. Is that okay with you? And you're like, yes.Link

And then you imagine, maybe for a real estate agent, what you start doing is, as opposed to organizing a meeting, you're more like, in the morning, I'll search for new properties in your markets that match your current clients. I know who your current clients are, because they've sent you emails telling you what they're looking for, and you have a recording of the onboarding call that you did with them. So I'll just do a first pass at the market every day, and then you just tell me the properties that you think would match. And then what I'll do is I'll send the email to them with the properties. And then if they say they're interested, I'll find a four-hour block for them to visit the property. But also, do you want me to add a property that's slightly worse and slightly more expensive than the one you're showing, so that they have comparison points?Link

LubaSo being super proactive.Link

Jean-DenisYeah. And I think that's, as I say it, it's crazy. No real estate agent is getting AI to do that. Not even the most AI-pilled OpenClaw real estate agent. But I think we're like 12 months from that with our product, out of the box. And I think we can do that for every vertical of job out there. So that's the way I think the tech can go. And I think if it's done well, you're never playing with a skill. You're not doing triggers. None of that... That's in the background, what's happening, but it's not, as a human, how you experience the technology.Link

LubaSo obviously there is a ton of different use cases. You just mentioned some. How did you guys pick which use case and which customer to focus on initially?Link

Jean-DenisYeah, we do everyone.Link

LubaYou do everyone?Link

Jean-DenisYeah. So, you know, what are the secrets? I'll give you one. I think people over-index on recent experience. Right? Over the last 15 years, people have built vertical SaaS companies. Vertical everywhere. Vertical, vertical, vertical, because it always made sense, right? If you're in a vertical, you could build a slightly better product for that vertical. It'd be easier to market, because you have fewer people to go after. The value prop could be more specific to your users, so they would like it more. You could probably charge more, because you have more vertical-specific features. So everyone's like, oh, who's your ICP? Oh, is that for real estate agents? Cool. You're the assistant for real estate agents.Link

And I actually think with AI, we're going to build broadly horizontal products where AI is making it feel vertical-specific for every vertical. Because at the end of the day, the use cases are pretty general, and the tools are very similar. You need to surf the web. So there's some building blocks that you need for everything. You need a table construct. You need a way to ask questions of the user, right? You need a way to surf the web. And you need to put these things together with the right goal and the right framing. But I think it will be possible to build a horizontal product that feels highly, highly vertical.Link

And I'll give you an interesting example for us: parents. So it's a work product. A lot of people use it for work, but actually a lot of people use it for personal life stuff. And one of the personal use cases is parents, because parents are always scheduling things for their kids, blah, blah, blah. And so if the product does scheduling really well, it's not just helpful for a salesperson. It's really helpful for a parent. But the parent, you know, they want slightly different things out of it. So when you read emails for the parent, what you need to do is know the names of their kids and what grade they're in, so that when you read school newsletters, you can just pull out the things that are relevant to the kids' grades. Right.Link

And maybe the way you show the data to the user is slightly different in the parent and in the work use case. Maybe the default sharing settings are different. You share more with your wife than you would with a coworker. So I think that's where all the nuance can be done. But I think the core functionality of the product is not that different. Yeah. So that's one of the things we believe in. It just means the way we are building the product and the data platform, and how we think about suggestions, is just not very vertical at the moment.Link

LubaYeah. What comes to mind with parents is, you know, Claire Vo's podcast, *How I AI*.Link

Jean-DenisYes, I love... yes. Out of all the AI influencers...Link

LubaYeah, she's great.Link

Jean-DenisShe's the one. I think she's the one that doesn't oversell. I think a lot of people are trying to get you to believe that Claude Opus 4 does the best thing ever. She's very concrete. It's great.Link

LubaYeah. But the reason why I bring her up is because, in my research of OpenClaw, when it was really hyped and everyone was talking about it, she did a bunch of things with OpenClaw, and one of the biggest use cases for her was scheduling for her kids, for, I think, 2 or 3 kids. And, like, one has a basketball game, another has something else. And the OpenClaw would serve as this house assistant for her and her husband. So that sounds like a very powerful use case.Link

Jean-DenisIt is. I mean, that's like... Yeah. Yes.Link

LubaDo you use it for your family?Link

Jean-DenisI do use it for my family. Well, mine are probably still different.Link

LubaYou don't have one for your family?Link

Jean-DenisMy wife and I both have a Townee assistant, but they can fully share information with each other. And so you get similar effects. But basically, if I have to ask a question of something that I know is in my wife's calendar or inbox, I can get the answer without having to ask her. Or if I need to organize something for the Parent Teachers Association, and I know she's super involved with that and some of the data is there, you can pull from both sources.Link

LubaSo it sounds great that it's a very horizontal product, but with a small team you still gotta focus somewhere, I assume. Or focus with marketing, focus with your landing pages. How are you figuring out where to focus or not?Link

Jean-DenisI mean, I think marketing and landing pages, you can run a bunch. You can use AI to really verticalize. But yeah, we have focus. So, you know, I always say this. The mental framework for me for most good product work is, you have accelerators, or differentiators. It's a bucket of things that if you crush, you are ahead of the competition. So we have two in that bucket right now. Two things that I think, if we do them world-class, get us ahead of OpenAI, Gemini, Spark, Claude Cowork, and so on and so forth.Link

LubaWhat are those two things?Link

Jean-DenisNo.Link

LubaOh. Sorry.Link

Jean-DenisYeah, yeah, yeah. Okay, so. And then there's table stakes, which is, if you don't do these things decently, then the market gets ahead of you. And then the last bucket for me is foundation, which are things that you could not do right now, but if you don't do them for too many months in a row, then you slow down and move really slowly. So we always try to have 1 or 2... With, you know, a small team, less than 20 people, we can only afford two things in the accelerator bucket. And those things you can't miss, you know. You have to do those, like, A+. So that's the way it looks right now.Link

So hiring... I mean, the reality is we're growing so much, both on user count and revenue, that I think we could use capital as a weapon against others. Meaning, you know, we can hire people to do really high-ROI work. Usually the problem with horizontal products is you get very different feedback from different people, and it makes it really hard to prioritize. I think for us, what's interesting is, different people that use the product still have fairly similar needs. And so if you do ten verticals right now, it doesn't feel like we have to build ten different features. It feels like we have to build three that work really well, and it can bring value to all ten verticals.Link

LubaI feel like so much stuff is happening here. Every time I'm here, I'm like, oh my God, I feel like I'm not a part of this zeitgeist and bubble, and I feel a bit of FOMO.Link

Jean-DenisWell, I don't like to call it a bubble.Link

LubaOoh. Oh, don't do that.Link

Jean-DenisDon't do that. No, no. But zeitgeist is better. But it's definitely... it's more back than it's been for a long time. It feels to me, it's the same energy that I think 2008 to 2013 had, or, you know, '97, '98. The only difference is, tech feels more heavy than it ever has. When I grew up, people were by and large very optimistic about all technology, naively so. Right. And I think we saw with social networking that it's not all, you know, kittens and puppies and roses everywhere. So I think you still feel the weight right now of, is tech good or bad for the world? And it's a little too heavy.Link

But outside of that, I mean, the energy is fantastic. And the talent density, because I think AI is so driven by the researchers and the labs, and there are only a few hundred of these people in the world. And they're mostly... I mean, there's some in China, and in Europe, but it's mostly here. And so then everyone wants to be around them in some way, shape or form. Right. When OpenAI's researchers were in the Mission office, people would want to be at the coffee shops around the office just so that they could overhear and run into researchers, you know? So, I mean, it's a little weird to think about the density that way, but I think it has moved the center of gravity towards here much, much more than in a really long time. And it's, you know, pros and cons, but I think if you have the luxury of being able to be here, right, it's a good place.Link

LubaIt feels really good.Link

Jean-DenisYeah. And it still has... People underestimate this about Silicon Valley, but people are very willing to still share information with others and be helpful. So I don't think everyone's looking over their back because someone will steal an idea or something. So it's the best of both worlds.Link

LubaYeah, I agree. So I'm hoping that also with the interviews I would be coming here more often, because a lot of stuff is happening.Link

Can you repeat again what that framework was? The accelerators...Link

Jean-DenisYeah, you need accelerators, table stakes, and foundations.Link

LubaAnd where did you learn this framework?Link

Jean-DenisNah, it's pretty common. I don't know.Link

LubaOkay. Got it.Link

Jean-DenisYeah. I don't even know where it originally came from, honestly. But when I was at Plaid, when we did strategy with the exec team, the CEO was very good about trying to be like, what's one thing that, if you look two years out and we do it better than everybody, the competitors can't catch up, if you do that thing right? And those are the accelerator category. It's hard to figure that out. And often you think it's an accelerator, but it's just a table stake, because everyone else does it. So it does mean you need to believe something slightly different about the world. You need to have an opinion that's not common.Link

So for Plaid, one early on was... Plaid allows you to connect your bank account to other apps. And so when you use Plaid, the UI that you use to connect your bank account, it's owned by Plaid. It's a Plaid widget. It has Plaid branding and all of those things. And Zach was very clear, as was William, the co-founder. He was like, if people get used to seeing "Powered by Plaid" for three years, we're going to have higher conversion and more trust. And that means the newer fintech apps will want to use Plaid. They won't want to use someone else's, because ours will have been there. And all the competitors, to get customers, were allowing people to really white-label the experience. So, you know, Finicity, a competitor, never said "Powered by Finicity." It just looked like the app.Link

And so when Plaid did this originally, we lost one of our top five accounts over that decision. And that's risky. That was a lot of revenue at that point. The top ten customers were more than 50% of revenue. So if you lose a top five account, you're losing... But he really believed in that. And obviously it ended up having huge network effects. I mean, once we had that, we were able to make an experience for when someone comes back the second time, where we're like, we recognize you. You don't need to put in your full credentials to connect the bank account now. And that was a huge boost to conversion. And then that was even more of a reason for people to use us, which means more apps would use Plaid, more users would come through, we'd have higher recognition rates, higher conversion, more like that.Link

But to do that, you have to be willing to fight this war on branding. And every startup knows, if you go white label versus not, it's really hard early on to win that. And the fact that he had that conviction... So anyways, to go back, that was an accelerator for the business. If we were willing to power through that narrative and get our brand to be everywhere, eventually we would win. And when your sales team is not going to hit quota because the customers don't want the... want the white label, you got to have a CEO who's willing to say, that's okay. I will make up for your bonus in some other way, right? But you don't want it to become an excuse, so you have to be careful. And I think he was excellent at doing that.Link

And so, you know, when you're an early-stage startup like us, when I ask myself, the stakes are not so stark. But there are some things, which I won't talk about, that I think, if we do really well, that are against what other people are trying, and risky... If we don't get them, then maybe we're out of luck. But if we get them, I think it really puts us in a little bit of a different space in the universe. And some of these things are obvious. Like, you name the assistant, it has an email address, it has a personality, it's kind of playful. That's counter-positioning, but that's copyable. But there's other things that we're trying to do that are pretty hard, I think, for people to copy.Link

LubaSo are you worried at all about the labs doing something similar, you know, OpenAI...Link

Jean-DenisYeah, whatever. I'm worried about everybody. Facebook, Google, the labs. But worried... I mean, what can I do about it? I can't go into the lab and poach their entire product engineering team. So the only thing I can do is believe that I'm building something that has a different direction. Because, look, we're building a way for knowledge workers to get more out of AI, right? Fundamentally, there's like a billion-plus knowledge workers in the world. Most of them use AI like three times a day, like search. So the market, I feel like, is very open.Link

I don't think I'm going to get 90% of the market, or 80% or 50% of the market. I just think there's going to be a lot of nice, really big, juicy islands in the market that are going to be worth it for us to conquer, if we build an opinionated product that some people are like, I like that, I want that, I like the way that tastes, I like the way it looks, I like the way it makes me feel. I want that. Not everybody has to like it. That's fine. Just enough people that we're a really meaningful business. And if I can build something that everyone likes, that's nice too.Link

So when I think about the labs, and when I think about Google, I think it's kind of theirs to lose, for Google. They had everything. They have all these people using Gmail and Calendar. And so, you know, maybe they retain everybody, and everyone that uses those products just decides to use Gemini all the time. Or maybe there's actually an opportunity there to build something different than the Google-flavored version of this product, and convince, you know, 50 to 100 million people that this alternative is really worth it. So that's why I can't fundamentally worry about it. Like, let me try to build something unique that's valuable, and then the market will tell me just how valuable it is.Link

LubaFor sure. One of your former colleagues slash angel investors, Christina from Vanta, said that back in the day, you made Team Cash at Dropbox really cool. Like, it was super uncool, and then you made it really cool. What was the story there? And two, how are you making Town cool?Link

Jean-DenisWell, okay, so I don't think... Okay, so at Dropbox, when I joined, I was like engineer 100, and I was not, at the time, the best engineer. And so, you know, all the really good people who joined Dropbox, they were going to work on file storage. And I was definitely not credentialed enough to work on file storage. So when I got there, basically, in my onboarding class, which was like ten engineers, they put all the new people in a corner of the office, in these empty desks. And then every day, a couple of people would disappear, because they would be poached by teams. And then eventually I was literally the only person sitting in the newbie area at the end of my two-week class, and I was like, I guess nobody wants me.Link

LubaYeah.Link

Jean-DenisAnd I think they literally forgot about me. But they put me... They were like, we're creating a new team. It's gonna be a growth team, and you're going to be on it, and you're gonna be the first engineer on it. And I was like, cool.Link

LubaThey're like, where to put you?Link

Jean-DenisYeah. And I worked with the PM who ran the team, and his name was John Adams.Link

LubaOkay.Link

Jean-DenisAnd he was in charge of money. So it was very good. So it was John, myself, Chun-Li, designer Puck, this PM Mindy... This group of people, we all started to work on growth. The team kept getting new people in, and we were basically trying to get more people to pay for the product. And eventually we were trying to get more people to pay for the team product at Dropbox. And I don't remember how, but the manager of the team, this guy Dan, decided that we would go buy tracksuits. Red tracksuits, like Euro-style tracksuits. And there's this guy Dimitri on the team, who's Ukrainian, who had a natural propensity for tracksuits. And so every time someone joined the team, we would all go to the Adidas store and we would buy a tracksuit.Link

And I don't know how to explain it, but from the tracksuit, then we renamed the team Team Cash. And then there was an all-hands presentation where I made fun of Drew Houston and his credit card, from the original billing implementation. And I don't know, there were all these moments. And then eventually the team couldn't be Cash anymore, because they had to grow the business. So then we called the team Huge, and we had giant stuffies. I don't know, all these things happened.Link

And over time, inside of the company, every time we presented at all-hands, we would all wear our Cash suits the whole day. And then people knew that at all-hands, it was going to be like, we'd beat revenue in some way, because the only time Team Cash would present is if we'd crushed some A/B tests that would give us more money. So it just became a little bit meme-ified. Yeah, but I think the manager of that team, this guy Dan Wheeler, I think he was the real... he started it. And then a bunch of us, randomly...Link

Yeah. So, okay, we got these hoodies. There's something called the cash cat. Like the Japanese cat. The prosperity cat. So we got these hoodies with the cash cat on it. They were super ugly, and, not in a bad way, but they're just not very... But they had the cash cat.Link

So there's a thing at large companies that I found, which is sometimes the best projects to work on are the projects that are important for the business, but that no one serious in leadership really cares about how it's done. Because then you get to get wins all day, because it's low-hanging fruit, because no one wants to work on it, because it's not cool. You know, at Dropbox, cool was the desktop client or file storage. Not cool was getting people to pay for it. But in fact, the board really loved money.Link

LubaAnd so for you to make money cool for them.Link

Jean-DenisThat's right. So you want to go on those teams, because as an engineer, often it means you get more resources. You can have wins. There's no exec that has a strong opinion about taking credit cards. So there's nobody who's coming in there and ruining your roadmap. But unfortunately, those things are not super cool. So you got to make them fun. So I think in that section of my career, and then I did this also a little bit at Plaid, I love making things that are not cool feel fun and quirky and cool. And you can do it by being a pirate ship, but I think that can be a little feel-bad for everyone. I like to do it in a way that can be broadly celebrated. So yeah, that was one example.Link

And I'd also... I like dressing up teams. I think it's super fun. When I was at Plaid, we had a team that was building the bank integrations, which was literally the blood of the company. If that team wasn't the best, most fun team, and people didn't want to work on it, it'd be a big problem for the company. And we had an office in Salt Lake City that focused on that. And the manager of that office, I always remember, I was visiting, and she's like, oh, we're doing a culture-building thing. And she had us all wear colorful tights, because she would always wear really colorful tights. So there's a bunch of us visiting, and we're wearing literally these rainbow tights all day in the office. Very weird. So I have these photos of us doing a thing like the Rockettes, you know, with the tights. It's just amazing.Link

Anyways, I like making things fun. For Town, Town is kind of fun on its own. We have the characters, the Townees. And so I think naturally, when you're 15 people, 20 people, you don't take yourself too seriously. You can have fun. And we've made that a good part of it, naturally. But I don't think we've yet super meme-ified...Link

LubaThe team could start dressing up like the Townees.Link

Jean-DenisYeah. This is, you know, I think... I think it's less fun to be a CEO than it is to be in between. Because CEOs are crazy. You know, founders are always a little crazy, and then the execs below, their job is to take crazy and modulate it, so that, you know, the team can be like, okay, I can deal with this.Link

LubaExactly.Link

Jean-DenisSo I think I really like that role. I'm really good at modulating, taking stuff from above and creating fun underneath me. Someone else can do that now. But no, the problem is, now I'm the founder, and I think I'm just becoming a weird founder. Yeah, I don't know. My great fear is that I'm becoming the difficult founder, unbeknownst to me. But you can't... You're not self-aware.Link

LubaSo are there already some, you know, signs that...Link

Oh, wait. So why do you have this feeling that you're becoming...Link

Jean-DenisWell, I just have a fear. It's a fear. It's unfounded for now.Link

LubaAre there any examples that you've seen of founders who are very weird, and you're, you know, worried about becoming them?Link

Jean-DenisNo. Come on. We're not going to answer that question. Oh, they're all weird. Because, it's not their fault. It's because people don't say no to them. People don't give them as much feedback. I'll say for sure, even now, the people who've worked at Town who worked with me before, which is like two-thirds of the team, they give me shit all the time. It's fine. But new people who join, especially now that it's going well, who just joined, it's like, oh, it's the CEO, a little bit more.Link

No, I mean, I try. I'm approachable. I go on walks, I try, but there's this natural little gap that happens. I don't know. Did you see the snapshot of gases yesterday? I mean, maybe it's amazing and I look forward to using it. But there's a part of me that feels like someone should have said something, you know? So I think it's more like that fear. We'll see, we'll see.Link

LubaWhat's your CTO personality versus founder personality? What are some differences that are already obvious?Link

Jean-DenisHmm. Well, oddly, I think I'm spending more time on technical matters as founder than I was as CTO. I think as CTO, I felt more like my job was to...Link

LubaAlign.Link

Jean-DenisYeah, and let the staff engineers do their staff or principal engineer thing. I think now I'm more like, oh, that's not the way I would do it. And I can't risk the company not having the right technical decision here. But honestly, I'm now catching myself, because I just don't know that much anymore. And all the engineers on the team are better than I am, so I gotta listen. But I think I am more forceful, and I think I'm less aware of the boundaries than I should be otherwise.Link

LubaHmm. Interesting.Link

Jean-DenisBut look, we're young, so I have time to become a crazy CEO next time. Yeah. First, we need to be more successful. I think it's only cool to become these personas, by the way, if you're very successful.Link

LubaYeah. Who are the CEOs that you look up to and are trying to model?Link

Jean-DenisI don't look up to people.Link

LubaOkay?Link

Jean-DenisI think it's a bad habit.Link

LubaTell me more about that.Link

Jean-DenisI mean, you don't know the people...Link

LubaYour father, or your mom, or... Look up to someone.Link

Jean-DenisYou know, I used to know the full picture. I think looking up to anonymous people is a little bit dangerous. I'll tell you. I'll give you two answers. One of them is, I love peers more than I do mentors. So I love to have deep relationships with people who are going through similar things that I am, right? So when I was at Plaid, one of my favorite people and a really good friend was this guy Tido Carriero, who at the time was CPO at Segment, you know. And the companies were basically growing at a very similar growth rate for a lot of our lives. So we were just comparing notes. What's going on? And emotionally, it's really nice to have someone that's in the same cohort that you are. And so right now, it's the same. I have a few founders that I like to connect with who are going through the same thing, and I like that relationship better.Link

And then on heroes, I mean, the only... and I don't want to... this is going to sound weird. I ask myself, you know, every age has a few people who are bigger than the age. So I think as a technologist, you have to look at Elon Musk, despite all the faults.Link

LubaYeah, for sure.Link

Jean-DenisBecause there's a lot of other people that are successful where my vibe on it is, well, I never had the insight that they had, but if I'd had the insight, I could have done that job. And I think with Elon Musk, it's way too much of it. I was like, I never would have done it. There's no way I could have done it. It's not one thing that I couldn't have done. It's like 100, you know. And really, there's a lot of things in the world where, if it wasn't for that person at that point in time, well, it still would have existed. And I think with him, you're like, I don't know if we would have as many rockets in space right now. That might have happened ten years later. And so it's hard not to look up to that.Link

And, you know, I love history. And so, you know, I'm French. So when you're French... I'm American too. Don't worry.Link

LubaPeople with two nationalities.Link

Jean-DenisWhen you're French, it's hard not to read a lot about Napoleon, right? Because he's done so many things. The French legal code is from Napoleon and all that. So anyways, you look up to Napoleon, and you're like, the guy's nuts, for sure. No doubt. But in his time, lots of people didn't like him and had no respect for him, and they were all small-minded people at the end of the day. But it's okay. That's human nature. Human nature, it doesn't matter how good you are or how bad you are. 51% like you, 49% don't like you. It's not much different than that.Link

So the thing about Elon is, he gets a lot of hate. And obviously, you know, he asks for a lot of the hate. But I think he's a large-shaped human in the history of the world, and it's pretty cool to witness that. There's other large-shaped people in our life right now. But yeah, I still think he's got a lot ahead. I can't look up to him because I'm not him. And I think there's too many negatives and words that I just have trouble identifying with. But it's just, the body of work is one that you would compare not to other people living today that much, but to people living over the last couple hundred years. I think it's that level. There's, you know, like ten, 20 people like that.Link

Luba100%.Link

Jean-DenisYou know, over that period. So that's pretty cool to witness, for sure. And the nice thing about the internet is we know a lot more about him than we would at other periods of time.Link

LubaYeah, than about Napoleon, probably. How did you guys pick your Series A investors?Link

Jean-DenisThey picked us. Ben Horowitz signed up for the product. And then he was like, hell yes, my Townee is the best. I want to... Basically, him and Alex Rampell just became power users. And then Alex texted me. Alex Rampell had been on the board at Plaid, and he just texted me, and he was like, we'd like to work with you. And, you know, I had a list of 3 or 4 investors that if they called, I would say, let's take this seriously. And they were definitely on the list. So it wasn't that difficult.Link

And then the other big investor in the Series A was Forerunner and Kirsten Green. And she's been awesome. And she has a lot of prosumer, consumer experience, which is a big part of our product. And so I felt like those were two really good partners to have behind us.Link

LubaFor someone listening, you've obviously worked with a lot of investors in your career. With Plaid, I've seen people on the board, now with Town. What's the best way to leverage...Link

Jean-DenisI don't know. Yeah, I'm not good at asking for help, so I'm not the right person to ask. I feel like I under-leverage our investors. So I think I like investors that are forceful and tell me how they can help me. I don't think I know how to leverage them.Link

LubaIf they're watching. What type of help are you looking for?Link

Jean-DenisWell, we're gonna have a board meeting in a month, where I have one... I think, like a... Yeah. I don't think I can talk about that one here.Link

LubaOkay.Link

Jean-DenisI mean, I go back to the peers. We have to make some product decisions right now that are not unlike product decisions that companies like Harvey or Legora or Decagon or Sierra have had to make, for example. And where I expect the board to be helpful is both giving me their opinion, because they've seen similar decisions made at those companies, but also putting me in touch with those founders, to get not the fake answers, but the real, like, hey, be honest with me.Link

Otherwise, I don't know. Alex Rampell has helped close seven out of our last eight candidates.Link

LubaWow.Link

Jean-DenisAs he likes to remind me in text messages. But I mean that in a very positive way. I think recruiting and candidate closing is... I think a great investor would really help with that. You know, when you're early, you don't have that much legitimacy yet, right? But you need to close incredible people. And so, you know, if you have someone who has a real track record as an investor saying, I don't write many checks a year, I wrote this check, this is why I wrote the check, this is how we think about the opportunity, that's very meaningful.Link

If you have a good answer on how to leverage investors, let me know, because I think I could do a better job there.Link

LubaWell, I actually think it's just asking for help, and being like, they're there to...Link

Jean-DenisI know. I'm not good at doing it.Link

LubaIt's a very obvious answer, and I don't think there's rocket science there. Everyone that I know whose investors are so helpful are very ruthless with being like, this is what I need. Because, you know, investors have so many people that they work with, they don't really know exactly how to prioritize their time. If you're not...Link

Jean-DenisYou have to be demanding.Link

LubaDemanding. Yeah, exactly.Link

Okay, I have two more questions. What are you struggling with as a founder personally?Link

Jean-DenisStruggling with... I mean, the stress and the hours is a constant battle. Being... We talked about it before. I wonder if we should be hiring and growing more aggressively than we are. You know, I don't want to get too much over my skis. But I have a good friend and peer whose point to me is, look, with the kind of product-market fit that you have, the way you should think about it is not just the next move, logically, but if you do the next three moves now and everything works like you expect it to, how much more successful can you be for the three moves after that?Link

Whereas I'm much more like, cool, when the metric gets here, then I do this. The metric gets here and I catch up to it. And he's like, go way ahead. Because I don't, you know, I don't want to wake up and have over-hired, overspent, overbuilt. I don't want to do those things. And he's like, the problem is, AI is moving so fast that if you don't always assume that it's going to work out, you're going to be behind the 18-year-old founder who just YOLOs a little more than you have. So I think I struggle with the right risk framework.Link

And then, I mean, I have strong opinions about things. Like, we're five days a week in office in SF. For a while I was like, I love that, because I love working with people. I love whiteboards, you know. But we have a great network in New York. And so I was struggling with it, and we were just like, let's hire in New York. Now we have like a four-person New York team, like, overnight. And I just feel like an idiot, because I should have just made that decision earlier. So there's that constant level of struggle.Link

But otherwise, honestly, today... If you'd met me like seven months ago, it'd be a lot different answer. But I think right now, the way someone put it, someone said I exude the kind of enthusiasm that you generally get out of a 25-year-old consumer founder who's found product-market fit. So I feel like I'm in a pretty good place.Link

LubaThat's fantastic. My last question: being a founder is the ultimate personal development journey. What are you learning about yourself?Link

Jean-DenisOh. I don't know if... That's a hot take right there. I don't think everyone would want it or would consider it the right kind of personal development. It's very selfish, and it's very isolating.Link

LubaI'm not saying it's the right one. No, it is one. It is one.Link

Jean-DenisYeah. Okay.Link

LubaSo what are you learning about yourself? Anything surprising?Link

Jean-DenisYou remember? We've got to go back to Marc. I don't want to introspect too much. I might find some things I don't like.Link

LubaThat is true.Link

Jean-DenisNo, I keep moving. Yeah. What am I... Let me ponder that one for a second. It's the one that we've been talking about down this whole street. I think, you know, we talked about confidence at the beginning, and I think I have confidence that I can do this, but I think I need to be more overconfident than I am today if we're really going to win. And finding that, and making it my own, and making it feel personal, I'm still working that out.Link

LubaWell, hopefully this is your one takeaway from the conversation.Link

Jean-DenisIt was really good. Yeah. It's therapy. This is therapy.Link

LubaYou just got to it at the end. Jean-Denis, thank you so much for your time.Link

Jean-DenisThis is awesome. It was awesome seeing you.Link