Apply

Podcast guests

Ali Ansari

Ali Ansari is the founder of micro1 and a guest on the Solo Founders Podcast.

Episode 33

A conversation with Ali Ansari

Published · 1 hr 8 min

Ali Ansari never had a co-founder - he didn't know they existed. In this episode, the micro1 founder tells Julian how he went from selling his dad's shoes on eBay at 12 to deleting his profitable AI recruiting business and going all-in on training data for frontier AI labs.

Transcript

Transcribed automatically, so names and words may be wrong. Each time opens the video at that moment.

0:00Ali Ansari: Why do you need a co-founder? I couldn't come to any good reason. Frankly. I just thought you would just start a company and you're running it. We lost about 60% of our revenue overnight, and we were, like, very close to actually not surviving that. What we do at micro is we build environments for labs to train their models. We have 150 companies that are signing up every single day right now to partner with us.

0:18Julian Weisser: What's the case against solo founding?

0:21Ali Ansari: If you're building a really difficult company, then there's potentially a case for co-founding with a couple of folks that have these different expertise that are complementary.

0:28Julian Weisser: What's the case for solo founding?

0:30Ali Ansari: You can make a lot faster decisions. There's sort of one person like clearly leading, and I think there's a lot of value to that. If you don't make these, what I would call kind of like outlier decisions, you're gonna have a pretty like normal return, but you're not gonna have an outlier outcome. Sometimes I hear from some founders that are starting something and they're like, you know, going through a process of like co-founder matchmaking. And I'm like, are you talking about just start the thing, right?

0:52Julian Weisser: I'd love it if we could just talk about the first thing that you sold online, sort of in sort of what that felt like and sort of what that meant to you at the time.

1:00Ali Ansari: Yeah. Yeah, absolutely. This is a yeah, it's been a funny story. I, um, I, I, I was, I think it was maybe like 12 years old, 12, 13 years old. And I was, you know, using my phone a ton, just like constantly, you know, got a new phone, was basically constantly on it. And my mom, uh, said something which was, you know, very simple, which is like, hey, you're on your phone a bunch. Why don't you try to make some money on it? And I was like, okay, great. You know, let me try. Sounds good. You found out what eBay is. And I was like, okay, we can, you know, kind of use the the automated listing process that they have. I could just go on there and, like, list anything I want and try to sell it. So I did exactly that. Uh, and I actually listed my dad's shoes as the first, first ever, uh, you know, item. And I remember, I, you know, I took his shoes. It was like some, some brand of shoes. I don't remember exactly what it was, but I just put some, like, arbitrary price. And the price was well under what what it was worth. So it sold right away. And then I and then I went to my mom and dad. I was like, this is like super easy. I just made, like, was like 70 bucks or something was like, oh, I just made like $70. Like, why is why is making money so hard? It's actually like, uh, that was like very frictionless. And and then my dad's like, why are you, uh, why are you selling my shoes? Like, so.

2:17Julian Weisser: So this was not this was not a shoe that you had gotten permission to say.

2:21Ali Ansari: No, no, I just like, randomly took it and listed it and then and then it sold. And then I realized, you know, very, very obviously that, like, I can't, like, list everything around the house we're going to run out of, you know, the house inventory. Um, and so I, I, you know, quickly learned, learned the concept of like, okay, I gotta, you know, figure out, um, things to buy and then and then and then resell. So I started going to garage sales and basically just buying stuff and, and kind of flipping them on, on eBay. And then my, uh, the thing that I kind of found a little bit of a niche in was, was textbooks. And the reason for this, uh, niche was I would like to grow sales, and and as I would bike, I would carry a backpack. And, you know, the thing that would fit well on backpacks was, was textbooks. So I would just focus on buying and reselling textbooks at, uh, at garage sales. And this was kind of my first, uh, you know, first steps towards the entrepreneurial journey, which was like very random. And I, I always look back and I'm like, this very random event could have not happened. And I'm so grateful that it did because it just like started the the process of learning for a decade before, you know, before the first, first thing ever, ever started working, which is, you know, which is micro.

3:29Julian Weisser: So and you said you were doing this, this was something you started maybe when you were 12 years old or something along those lines. What did you have any sort of entrepreneurial kind of or like business mind, do you feel like before that, uh, or is that something that you literally it was more of this thing with the internet and kind of being exposed to it and getting sort of turned on to the idea and then having that initial sort of reward that ultimately drove you to realize, wow, this is really fun. This is exciting.

3:55Ali Ansari: Yeah. So I think I'm realizing in retrospect that that, like, there probably is a little bit something in terms of like entrepreneurial kind of spirits that, that, that, that perhaps I have innately and I think it comes from, um, it comes from my parents, my, uh, my dad actually, uh, in Iran, he, he ran, um, a kitchen remodeling company. And he had, you know, like 20 or 30 employees and, uh, basically would design and build kitchens for, uh, for newly built houses and, and then also my, my mom, uh, when we moved to the US, she basically right away opened up her own daycare and she would take care of kids. And it was like, you know, small kind of local business. So in retrospect, I realized that, like, okay, maybe there's maybe the family's kind of, you know, like sprawling businesses. And then my, my grandpa as well, he had a he had a company in Iran. Um, but but that wasn't like a that that wasn't clear to me. Uh, back then at all. I mean, like, it sort of becomes a little bit more clear over time. But I do remember before, before this, like, garage sale, you know, reselling on eBay endeavors. There was, uh, there were some like the typical, like, you know, selling stuff at school, lemonade stands and trying to flip, uh, you know, buy cookies and go, go to, go to door by door to the apartment building that I lived in and just, like, sell cookies at the door. So I did a bunch of random stuff like that. But the, the, the books for that garage sales was like the first thing that was like, okay, here's a bit of a niche and maybe I can like build this into like a quote unquote startup. So that was the that was the first real thing I would say.

5:29Julian Weisser: If you're a kid nowadays, you know, 12 years old, what do you what do you think? I mean, do you think, do you think that starting on eBay is still like, is still the thing selling things that have already been made? I mean, obviously it's so much easier to build things now, but it feels like so many people I know, and you probably do as well, uh, got their start with eBay or, you know, selling things, uh, you know, selling sneakers online, that sort of thing. I'm curious, like how you think about that now. If you're a kid starting out.

5:56Ali Ansari: I think there's a huge value in selling anything. This is like when when people ask me like, what's your advice to starting a company? Should I go and like form a, you know, LLC or do these like, random things? I'm like, no, no, just get somebody to pay you for something that, that that you're selling. Like, just make, make exchange hands with someone with the product or service and get some money as a result of that. There's there's a lot of value in, in um, in getting to that point and there's a lot of learnings in getting to that point. So I think today

6:31Ali Ansari: the notion of selling something is probably, you know, maybe it's better to kind of focus a bit more on software because it's software's, you know, a lot more, uh, a lot easier to build now. Um, maybe it's yeah, maybe it's like digital goods instead, but but just going through the motions of getting someone to pay you for something has tremendous learning opportunities. So that's what I've optimized for. And it's not about like, you know, it's not about the money that you get. It's about the learnings that comes from, uh, getting getting someone to pay you for something.

7:03Julian Weisser: I would love it if you could take, uh, take the audience on this sort of journey from you start this textbook resale business, um, through to kind of the point where you were building the thing that kind of became micro1, uh, because I think that's a really interesting journey. And we could we could talk about, you know, a few things along the way there. But if you could just sort of draw that timeline for folks.

7:24Ali Ansari: Yeah, absolutely. So the first step was, uh, this, uh, garage sale effort, which was buying and reselling textbooks, garage sales. And then that became, um, I was trying to, like, scale up that effort. Uh, and basically I ended up developing this website called Cash Books Now, which, uh, would essentially allow the college students to go to the website and sell back their textbook instead of selling at a at a garage sale, and they would get a, they would go on the website. They would put the ISBN number for their textbook. They would get an instant quote of how much we pay for the textbook. And I would use Amazon's API to basically, uh, you know, give an instant quote based on the price of the book as well as the sales rank, which like, how fast is the book going to sell? Um, and this was the first like, you know, quote unquote startup that I, that I had, it was obviously, you know, very small scale in my room. The books would just be, uh, stored in my room. I had like a, I had a Sharpie that I, that I wrote on the wall with, uh, on the left is books that are not listed. On the right are books that are that are listed on on on Amazon and eBay. And so that was, that was like the, you know, the, the supply chain that, that existed. Um, but essentially this is my first, uh, you know, start up. I ended up selling a few thousand books that, you know, had a lot of, uh, learnings from it. Um, and then I ended up selling the inventory of books as a whole. And, uh, basically ended up kind of moving on, uh, from this. And it was, uh, you know, I was working on this for, for, for a couple of years. And then at this point when I ended up kind of selling inventory books as a whole, it was, um, I think it was around the time of sophomore year of high school, I believe. Or maybe. Yeah, I think towards the end of freshman year or early sophomore, um, year in high school where I ended up, uh, starting, uh, another company called Pure Link, which was basically a tutoring company. And the quick backstory of this is, uh, I, I was in this club called Science Bowl in high school, which was basically like a math and physics competitions club. It was like a you hit the buzzer, you asked very quickly. And it's, you know, it's fun competitions with other high schools and I my, my, my, my, my thesis was like, I think maybe others want to learn about this, like math competitions and like physics competitions and so forth. Um, so I took my friends in, in the science book club and we, we opened up a center in, um, uh, in, uh, in Woodland Hills, uh, on, uh, near, uh, uh, in near Topanga. And basically the goal was we get, uh, other students to learn from their peers about math competition specifically. And it turns out this, this market is very small. It's basically nonexistent. Nobody wants to learn math competitions. So we had to we had to kind of like, generalize the model. And it started to build, um, just like math tutoring in general. Uh, and the model continued to be this, like peer to peer, um, uh, dynamic, where the tutors were like a couple years older than, than than the students. So the students would it's where they want to be in a couple of years, and they sort of look up to them. And it was a it was a very nice dynamic. So that was my, uh, sort of second company. Um, and I ended up working on that for some time. Got it to, uh, roughly like six figure run rate in high school. And basically the model was, uh, Parents would subscribe to the platform to get access to, to a kind of a pool of prevented tours. And then we built this, um, at first was an in-person location, but then after it was like a virtual, uh, tutoring, you know, whiteboards and these, these sort of technologies that that allowed the two researchers to have. So that was my second company, um, worked on for some time. And then I had a, a very small, uh, acquisition on that. And then I, um, and then I started, uh, moon tech at Berkeley. Moon tech was, uh, and this is, this gets the, the origin story of, of micro1, which is essentially, uh, Montague was a software development agency. It was very, uh, very service oriented. Nothing too nothing too special. We basically just built websites and apps for other companies. Um, and one of the kind of, one of the limiting factors of this company was every time projects would come in, I had to assign engineers to these projects and basically like scale up the engineering team. So a lot of my days was spent, uh, interviewing candidates and basically vetting engineers and hiring them. And so I ended up building this, uh, internal tool that helped me vet engineers. And it was you could think of it as like it was essentially an AI screener that that helped me, you know, talk to the engineers and it would give me a report on how they did, and I would take those reports and use them to, uh, qualify the sort of like top, top candidates. That was, um, that was the origin story of micron, which is that tool became the, the version one of of what we were building at micron. And I ended up, uh, spinning it out and focusing on, uh, building the first ever kind of AI recruiter in the market, uh, which essentially allowed companies to source some of that talent for whatever talent they're hiring for, was mainly focused on software engineers, but but they could also vet other types of talents. Um, and that was, uh, that was sort of the beginning. And essentially the we also had built we realized that, like, we can use this tool, this AI recruiter tool to build a marketplace of talents. And there's, you know, similarities to like the Pure Link, the Turing Company that I had. Um, and essentially we started to build a really massive marketplace because people would be in an autonomous fashion, prevented, um, and then startups would, would be able to hire from this marketplace. So that that became sort of the model, uh, the sort of version one of the model. And we, uh, and then, yeah, we, you know, we got into a few million kind of in the, in the first, um, for first year or so. And then we, we pivoted the business, which we can get into.

13:12Julian Weisser: So there's so many places I want to take this, but I think the first thing is, um, is this concept of, uh, when to move on, when to stop, when to when to sort of pull the thread on one concept but leave a lot of it behind. Um, you know, I think that oftentimes people talk about people giving up too early or, or or maybe not going for long enough, but I think there's also just something to be said about, uh, you know, when to when to move on. And you've you've worked on so many different things over a relatively compact period of time. Some people you could imagine sticking with the tutoring thing for far too long. So what were some of the things you've learned about, you know, when to when to move on? What what were the actual these that you kind of talked about? Hey, like we got a small acquisition offer or something like that. But I think, practically speaking, a lot of people would probably try to figure this out themselves. Like, what should I move to the next thing?

14:04Ali Ansari: There's actually two, um, two very specific things here that, that, um, that resulted in moving on for each of the companies. Um, so for cash books now, it was it was a

14:18Ali Ansari: I remember very specifically that the main way that we would get college students to go on the website and sell back their textbooks was, uh, keywords on Google ads and basically like the Google ad optimizations. And, and if we were not able to get enough, um. If the cost was too high per user that was still back there. Book through Google ads. Uh, the business simply wouldn't work. And we we got to a certain scale where, you know, we had, um, we had a few people a day that would be selling, selling back their textbooks and, and, and as we said, we would be as I, as I tried to scale the, the Google ads, it basically like it wouldn't work. The cost per acquisition would, uh, would wouldn't scale. And there was no and of course, there was like a ton of other experience with, uh, you know, other, other methods of getting folks to sign up. Um, which which also, like the economics ended up not working out. And, and I decided, okay, makes sense to let's let's, uh, let's just sort of I've been at it for a couple of years, you know, learned a ton. Let's sell the inventory of books as a whole and move on. So and, you know, there's probably, probably, probably could have kept going at it for, for a bit longer and uh, but but that was, that was sort of the catalyst for Pure Link and all these three, I think these three specifics, uh, will will allow me to sort of, uh, convey the, the sort of learnings from them. But, but, but, but, but the, the, the pure link was actually this one was interesting because a very interesting one because I, I was going, going out purely for, for too long actually. I, um, I started it sophomore year, uh, early sophomore year of high school and, uh, and I, you know, I got it to like a okay revenue scale, but then it was flat for, for a long time, I think it was flat for like over a year. And I remember very, uh, very specifically I had, um, one of my, uh, uh, one of my mentors, which is actually the, uh, my one of my best friend's dad. His name is, uh, Hamid Cohen. Uh, he, he runs a company called, uh, Legal Soft. And, uh, huge shout out to Hamid. He, uh, I remember I sat down with him and were just, like, discussing, uh, purely and like how things are going. And just like getting his advice, um, he has skill the company to like few hundred employees. And it was a, you know, great, great company. And he basically told me, like, right away he was like, why are you working on this? This is like, uh, it's not like an interesting, interesting thing to work on. And, and to me, back then I was like, this is the most interesting thing ever. I mean, how can you say this like, it's, it's, you know, it's the most, uh, it was it was my baby. And it was like the most interesting product to me, right? But, but but that moment made me realize that, like, okay, maybe it's not actually the most interesting product or business and like, maybe, uh, maybe, maybe there's there are other things that could be much larger scale. Um, maybe there are harder problems solved and so forth. Um, and, and I basically ended up at that moment of him sort of saying that I ended up it ended up resulting in me, um, going for the this small acquisition that, that, that, that was there and essentially just just moving on. And that was one of the best decisions I ever made. Because if I kept going with that, I mean, now the I, I mean, it could, it could, could be a cool company to build like AI tutors now. But but you know, before that it was uh, I think it's just hard market to be in um, and, and, uh, you know, just like overall a less interesting product because it's sort of the reality. So that was the catalyst there. And I think that the third is like for, for moon tech, it was very clear that like this tool that we built by accident can be a can be a company on its own. And it's and it just solved the really important pain point for me and a services company. Software development agency. Good. You know, there's sort of good good cash cow components to it. But but it's not really anything that scales up massively. But this tool could be so. So when I saw that catalyst is are great. Like we got to make them move fast. So I think all these three things they, they point me to to one one like major learning and of course, that there's a ton of like, uh, I think there's a ton of learnings, but like, there's one thing that's, uh, that, that, that would that I would emphasize, which is, um, the market really matters. And I think like super early entrepreneurs where they start their, their first company, they sort of they don't care about the market. They, they sort of ignore that. They just like build something, which, by the way, I think is, is actually probably the right approach because you got to just do something and like, you got to just go through some learning motions. But at some point you got to really decide the market and take take a lot of time to decide the right market. And, and of course, like, you know, our growth at micron has been has been great and we're very grateful for that. But also like we can't take so much credit like the market is just incredible. And we're so happy to be in this market. And of course we have to execute well. But but like the market has a lot to do with it. Um, so so that that is like a huge learning for me, which is like you got to be in the right market, uh, when you're when you're building a company.

19:22Julian Weisser: That transfer over from sort of what you're doing with interview attack and and then building sort of a marketplace of talent. Um, to actually building what micro1 is now sort of that transition. Some people might call it a pivot. I don't know if you would. It feels to me like it's an informed learning that causes you to sort of refine and kind of reorient the company. So maybe there is a pivot there. I'm curious how you define it, but the, um, the thing that's really like, really interesting about it is you are like, you're saying you're moving towards a better market because, you know, in some ways it felt like what I was reading about, sort of the history of the company when you were doing talent in that way and sort of the hiring side of things, recruiting side of things. Sure. Like that's something that could work. But is it the best use of the technology? Is it the best thing to meet the moment? And it seems like what you were saying just now is that, you know, you were if you could find a way to move away from bad markets and move into a great one, ideally, if it's not even necessarily ripping everything up and starting over, but actually taking some knowledge or something you develop and applying it towards an even better market. Magical things could happen. Is that is that accurate and sort of how would you how would you talk about that learning there?

20:36Ali Ansari: That's 100% accurate. And I think pivot is uh, is like a, a little bit of a wrong framing, which I appreciate you you framed it as like an evolution of the product we built. And it's like a kind of a learning of, um, uh, learning of a better market that this product that we built can be, like, applied to it. That's exactly what happened, which is we took the AI recruiter product, and we about a year and a half into building the company, we got a customer that was a massive customer rewards. It was like they became our largest customer within within about a week they were hiring like, you know, it was like 5 or 600 engineers. And we were used to like serving startups where we were like, they hired 1 or 2 people and maybe five, and if they hire five, there was a top ten account for us. So this account, you know, came in and and hire 500 people in like literally a week or two. And so we were like, well, you know what's happening here? What's what's. We were curious, like what? They're building this this has to be some, like, really crazy software if you need this, this many engineers. And of course what we're doing is training models and we're sort of going, um, we're transitioning, um, uh, in the data market from, from the generalist talent to expert talent to, to train models. And so we saw an opportunity because of this transition that was happening in the data market. This transition was, um, it resulted in a new kind of limiting factor, which was you need to recruit experts. And and we had built the perfect product for that, which is an AI recruiter that can recruit experts at the largest scale ever. Right. And this, this company that, um, became our customer was like a perfect point of proof for us that, like, this is this is going to work and we can we can definitely get into this market ourselves. And so we made the decision pretty quickly. Um, we didn't go all in right away. It was it was a it actually took longer than I, you know, obviously, in retrospect, we should have done it right away. But but we took a little bit of time. We took uh, it was actually like a probably almost a year, sort of like iteration towards going all in. And essentially the first step was like, okay, let's work with more of these data providers, which of course are now now competitors. But, um, at first we were just serving all the data providers that were just using our tool to recruit experts, and then they would serve the labs as their customers. Um, but then we started to realize that, like, we can build the rest of the infrastructure, which is, you know, well, well beyond the AI recruiter and build data platform, build our research team, you know, build our pipeline performance management system, all these different platforms that are as part of our, um, data infrastructure kind of play that we have now, uh, and just go and serve the labs and enterprises ourselves. And it's like too important, too important of a market to massive market to to not take this, to take this leap of faith. So we did exactly that. Um, and when we, when we took this, this all in kind of strategy, which again like took took a little bit too long. I remember I made this loom for the team and I said, uh, the title was like micro1 was going all in on data and um, and, and that resulted in us, uh, having to, like basically shut down the first version of micron, which was a lot of revenue as well, and just essentially like delete the revenue. And there's a lot of folks that were like, you know, are we sure about this? Like we're like literally giving up, like I think it was like 7 or 8 million in run rates. Like, this is, you know, we could continue to grow this, etc.. But but we had to make the decision of going all in and, and that was like, you know, that was the moment that, uh, massive inflection point for micron begun.

24:16Julian Weisser: So so did that that loom video that you made, did that come sort of in towards the tail end of that transition period where you were kind of iterating towards it, or was that more like at the beginning and you said, we're going to start this iteration? Just a.

24:28Ali Ansari: It was, it was, it was, it was the end of it.

24:30Julian Weisser: Very end.

24:31Ali Ansari: Like it was the final ending of the transition was like, all right, now we are all in. Forget about the forget about the AI recruiter. Forget about talent business. Uh, let's let's go all in on data. That's all we're going to focus on. And of course, everything we had built ended up becoming amazing internal tools for us to serve our customers. So so the the you know, the Zara AI recruiter to this day is an important part of our tooling. Um, but but there's like so much more now that allows us to serve our customers. But but yeah, that video was the last, uh, the sort of last step in the transition.

25:01Julian Weisser: So can you give a little bit of a picture of what the team looked like before you started that transition? Like how many people were there, you had it, was it a distributed team at that point? Because I think that something that a lot of people who I talked to, who are solo founders or even have co-founders, um, there are these moments in a company where you have a good amount of revenue. Things are in some ways going well, but there's an opportunity sometimes to go in a different direction or or take an insight. In your case, it was an anomalous thing, right? To have one company come on and look for 700 roles or make 700 hires in two weeks. I think you said that's that's remarkable. And I think that there's sometimes a temptation to, uh, to ignore those things or say, well, that's, that's an edge case, but you know, where things are going. Well. Over here. What what what would you say? Uh, the company looked like at that point and what was sort of the process of getting other folks on board and sort of, like, excited about this and also yourself. And in the process.

26:06Ali Ansari: It was a really difficult time, frankly. Um, because before we did this and this was this was a, you know,

26:14Ali Ansari: a couple of months before we made this, like, kind of all in decision, we actually went through, uh,

26:21Ali Ansari: a really difficult period where we lost about 60% of our revenue overnight. Uh, because of this one company that ended up kind of seeing that we might get into the space and basically, you know, ended up, uh, uh, terminating with us. Um, and we were like very close to, to to actually not surviving that. Um, and so this, you know, evolution of the company and going all in, it sort of became a little it became a bit personal, like, okay, we we must win in this space now. Um, because we, we almost lost pretty badly because of one of our future competitors right at that point. And, and so the decision was becoming very clear to a lot of, uh, to a lot of folks, But

27:12Ali Ansari: but for the broader company, it was still still not clear. And we and we had to just kind of peel off the Band-Aid and send the send the loom do, do sort of an emergency all hands and say, look, we're going all in on data. We're now data company. Um, and it was a tough couple of weeks for folks to understand what this means, you know, what sort of happening. But but I think these are the moments where and I, you know, I don't want to toot my own horn too much, but like, these are the moments where, like, the founder can only make these decisions and, and like, take the leap of faith and and and and and really like, try to try your best to convince a team to follow suit which, which is of course, a very important part of keeping the team intact. And if you don't make these, what I would call kind of like outlier decisions, you're you're going to have a pretty like normal return. You're going to have a you're not going to have an outlier outcome. I think an outlier outcome comes from these really like odd actions. The team is like. What the hell are you doing? Like this? Are you. Are you crazy? You know, this doesn't make sense if you don't do a bunch of those things throughout the company's trajectory. It's not even just early on. Like we need to keep doing things like this. Then I think you're going to build you might build a reasonably good company, but but you can't build a multi hundred billion dollar generational company. That's sort of what we're aiming to do. And of course some of these moves don't go well. There's a lot of examples of these moves that, that, that I've made that, that our folks in management have made that, that, that also like don't go, well, you gotta you got to recognize that as well. But but if you have, if you have a, if you have a few and you have enough that go well, you will build enough confidence to do more of these outlier actions, which I think are very important. Um, and this relates to um.

28:58Ali Ansari: This relates to, to to to to a point that, that I think, um, I sort of remind myself of, which is only the founding team can

29:09Ali Ansari: inject risk into the company in the in these ways, and you must inject risk into the company for for for for these sort of outcomes to be possible. And so, uh, I, I enjoy these moments. They're, they're, they're difficult. But I think it's what makes it fun. And there's another I'll give you one more of these moments, which was uh, uh, which was uh, I think, yeah, it was maybe also like two ish years ago, uh, where we, where we ended up, um, when we were using our, uh, AI recruiter, Zahra. We, we we had a a human interview process after as well. And a lot of cases, not in every case, but a lot of cases where Zahra would give a report and there was like a 15 minute human interview. And then so there was still a fundamental bottleneck of human time that was spent. I remember a couple of years ago where we had had this moment where we said, okay, the product is the product is not good enough. Let's take the leap of faith. and our recruiters no longer do any interviews. It was like overnight we we again, we did this like emergency all. Has everyone thought ever thought I'm going crazy again and I, I told like cancel all of your calls this week. We don't we don't do a single screening call after today. And this was this was like a decision that resulted in like there's a few folks that were like about to leave. And it was like it was like, you know, it was a it was a time. Rats. I had to make sure to keep the team intact. And this was another decision where our our expert pool ended up growing way faster. And if you look at the there's a there's a very you know, there's sort of a funny chart where it's like our expert pool growing very much matches our revenue growing. Um, and, and so if we, if we didn't remove this constraint, our expert pool would not be grown the same. And probably revenue would, would, would, would follow that. So there's a ton of these hard decisions that that must be made. And I think only the only the founding team can, can just make these with with conviction.

31:01Julian Weisser: Let's talk about solo founding in that regard then. because, um, I think that I read that one of the first companies you started, you did you did work with other people as co-founders, right? Is that is that accurate, or was it or have you always been a solo founder with everything you've worked on?

31:15Ali Ansari: Yeah, only sole founder.

31:17Julian Weisser: Okay, interesting. So what um, what what was it sort of that that kind of caused you to be solo like. Were there ever people who kind of pushed you to to have a co-founder or said you should have one? And I also realized that and this could be part of the conversation here too. You know, you were starting a lot of these things while you were still in school, in high school. Uh, and while I guess, like through college and, you know, at least what, you were getting started, you were in LA, you were in Woodland Hills area. Um, it's not a place necessarily where you're as exposed to, like the culture of Silicon Valley, which in some ways is probably a really good thing. Right? Because you're not you're not being indoctrinated with random things that you should that you should supposedly do that really, there isn't necessarily a reason to do so. I'd be very curious to hear kind of your perspective on solo founding. Yeah. We'll also talk about founding team because founding team is so important, especially for solo founded companies. But can you just share a little bit more about kind of how you've thought about being a solo founder or maybe not thought about it, considering it wasn't a thing that was maybe top of mind at any point early on.

32:20Ali Ansari: Is exactly the last thing you said, which is, I didn't realize it was a concept of like co-founders, even like when I started my first first, you know, a couple of companies when I was in high school and, you know, in LA and so forth, I, I just thought, you just start a company and you're running it and there's no, you know, I didn't know what a co-founder is. So, so and and yeah. So there's more of like a out of, you know, out of just not knowing I think. And and then when I, when I, uh, I remember the first time where I, you know, of course I learned about the concept, you know, later on. But the first time where I like, heard this concept really emphasized was when I was applying to YC for, uh, for one of the, one of the previous companies. And, and I think my understanding is like they don't say it explicitly on the applications or anything, but but like everyone sort of knows that like, oh, you sort of need a co-founder and they look for that. And I was so confused that like, why, why do you need a co-founder? What's the like? I was trying to reason through why this makes sense. And, and I couldn't come to I couldn't come to any um, any like good reason, frankly. And I the thing I came to is that that I think investors are trying to de-risk in terms of if one founder like quits, there's there's more that you can sort of rely on. So I think that that maybe could be a maybe a reasonable explanation on why certain investors and why she pushes for it so much. But, uh, but, you know, I don't think it's something that, uh, is, is required. But I do say this with, with um, I think for, for micro1, it's very important to note that our founding team is incredible. And I think like, I, I don't I don't I don't even consider myself a sole founder for micro wants to be honest. The founding team is so good that. And and you know, I there's there's a bunch of folks but but the one I'll name is like Will or Crow. He joined us a couple months in and he's, uh, he's been just incredible. And and also, uh, Joshua Browder, one of our early investors, uh, that put in, you know, first ever check. There are folks like that that are like, feel like co-founders are effectively co-founders. And so, uh, I think it's really important to optimize for people like that very early on. But I don't think you should look at it as like, I'm not going to start the company until I do like a found co-founder, matchmaking, all this like nonsense. I think, like, just just start the thing, get somebody to pay you something and and learn and then and then and then once you build, you know, the MVP or whatever, you start building it. There are folks that are really interested in potentially joining you and whether they're co-founder or founding team, whatever, whatever label you put on it, I don't think you should look at it as a constraint of starting. Like that's what pains me is like sometimes I hear from some some founders that are starting something and they're like, you know, going through a process of like. Co-Founder matchmaking. And I'm like, what are you talking about? Just just start. Just start the thing. Right. So that's the I think that that's the thing I would say. But but it doesn't I think like the, the the thing that you cannot, uh, the thing that you cannot kind of, uh, the, the way to misinterpret, interpret this is that that like founding team is doesn't matter and stuff like that. I think in fact, it is so important to have an amazing founding team. Uh, if you're if you're a sole founder, um, but but you shouldn't like, wait on a co-founder is my is my other.

35:27Julian Weisser: Totally. You know, um, my friend Eugenia, um, who's building a company called wabi right now, she talks about how when she started, she started her prior company solo, too. She thinks that there's this real beauty of of the solo founder and the founding team that that you don't have when you have co-founders necessarily, because there's sort of like the co-founder and CEO, there's the co-founders, and then there's the founding team, but there's so much more closeness between, you know, a solo founder and the founding team. It also allows the founding team to just have so much more ability to have influence within it, versus having that sort of maybe co-founder as like the middle layer or hierarchy type thing, which I think is, is is some ways maybe there's value in it, but in other ways it's, you know, it might constrain the founding team. I'm also really curious about, um, sort of as you were, as you were going through this process of essentially we call, I call it pulling the thread, like instead of instead of pivoting. It's like, it's like pulling the thread on an idea. I'm curious sort of what what it was like with the founding team and also kind of how you think about, um, you know, what it was actually like for them to go through this process to of kind of like consistently evolving, because I imagine there are some people who stayed from what you were building, sort of the recruiting like, uh, recruitment analysis sort of thing all the way through to micro1. Can you talk a little bit about that journey for the team and how it was like with you, sort of like leading people through and having them work with you through all these different transitions.

36:58Ali Ansari: Yeah, so

37:01Ali Ansari: very grateful for

37:03Ali Ansari: most of the folks that are very early. Uh, almost all have have are still part of their journey, even though the journey is like, you know, we've changed quite a bit. Right. Um, and and it's, uh, I think I think the reason is it's it's similar to, similar to what you were saying, which is like there's less of a hierarchy and there's like very quick decision making from the, you know, from, from the founder. And, and that decision making is, is as a result of really good and close debates that happen with, with the, the founding team. Um, and so, so the founding team is like very involved in all these decisions. So it's not like, you know, we're sort of iterating. And there now there's like a new, new thing that we're working on that they sort of just have to adapt to. They're they're part of the process of those iterations. And they're, they're they're a huge part of the process. So I think that's an important part. I think the other thing that I'm very lucky for is that, um, a lot of the folks from my previous company, Moon Tech, when we were doing the agency, um, actually ended up transitioning into, like working, uh, working at micro1. And there was a couple months there was like a few months period where where I was sort of running both before I shut down the agency. And so there's some folks like working at the agency and then, you know, some new folks at micro1. And then once we once we shut it down, most of the people that that, that I was working with at that at the previous company ended up, uh, joining micron as like basically the founding team and, and uh, and they're uh, yeah, a group of very, very, uh, special, special humans that are, that are with us to this day. And, uh, I think a big part of it is what you said, which is the, the closeness of, uh, of me with, with the founding team and the fact that they're super involved in all the decisions that we make. And it's really, you know, it's really our it's our company. It's not you know, they're all material shareholders.

38:57Julian Weisser: So can you say more about the sort of debate culture and sort of and also more just like company culture in general, how do you think about it, what you've learned, um, mistakes that have been made, like successes that you've had.

39:08Ali Ansari: I think there's a couple of things that we, um, really emphasize in our company culture.

39:14Ali Ansari: One is we've built a bit of a unique, um, remote by default and asynchronous by default culture, which for a company like ours, that's like, you know, say like pretty hard core, like, everyone works, like, super hard. You would not you would not think remote when you when you think of those, when you think of this industry and you think of the sort of, uh, the hard core nature of micro1, but, but we've, we've set up in a way where, where just it just works and it's like the, the, the, the, the, the joke that I make is like, you know, you can have you can have, uh, this, this 1996 stuff, which I think is like very cheesy and, and, you know, uh, yeah, very cringey overall. Or you cannot have those like arbitrary constraints and, and you can just sort of inspire the team to work weekends versus like forcing it. And everyone will work all the time because there's a lot of interesting problems to work on. And if and if somebody is working from their couch at 2 a.m. versus like being forced at the office, that's fine. Like they're probably more productive there anyways, right? So so we look at these like we question these arbitrary constraints. And we think questioning arbitrary constraints allows us to get, uh, really good talent that aligns with that. That could be global. It could be NSF location is basically not a constraint. Um, and it also allows us to move very fast because it we when we have discussions and debates, they're very targeted. There's no you know, we don't we sort of avoid the small talk that that that that might happen at a large scale at an office. Right. We, we open a thread with a very clear problem. The person that opens up presents the solutions that they're there thinking about. The only people that need to be involved are involved. We limit the number of people that are involved in discussions. Um, and we try we we try to close. We try to close the thread as fast as possible with what we call final actions, which are minimized. Minimize the number of messages until you get to a final action is what we sort of aim for. And so the company ends up being very action oriented. And of course like saying, you know, saying this idea of like being action oriented is easy to say, but but you have to be very specific about it. And the way that we're specific is, is, you know, what I just said earlier, which is you open a thread, you propose a solution, you minimize the number of messages, and you get to the final actions. And the person that opens the thread in most cases owns the final actions. And and the most, the most honorable thing you can do is, is act not not give an arbitrary opinion. And of course, a lot of opinions are not arbitrary. But if you if you believe your opinion is not going to add value to the discussion, then we avoid. We avoid giving the opinion. It's sort of our culture. Because if you. It's really fun to to have like long discussions about really cool, cool things. But but the ultimate actions that come as a result. If you if you if you ten x the length of the discussion you you will probably converge the same set of actions. So you might as well. You might, you might as well converge much earlier and basically like, you know, stop the discussion. Right. So that's one thing that we care a lot about. The other thing is, um, and and this this will sound a little bit cheesy, but but it's important, which is our culture is very hard core, but we're also very kind. And we emphasize this every day, which is like we move fast and things get very stressful, but we cannot forget to just be pleasant to each other and like, just have a good time working. And I think, uh, I think this is maybe underlined at, you know, sometimes, uh, and we make sure to, to really take this, uh, take this as a core, core principle. Um, so, yeah, I think these are the two things that come to mind.

42:49Julian Weisser: I'm really curious to hear about, uh, sort of the, the action oriented nature of this. How do you. Because ultimately there are lots of people coming from all over the world, all sorts of cultures, different company cultures, different places that they grew up. Um, how do you actually help people sort of go in that direction? Because I think that some people might want to do, but maybe they haven't experienced a company like that before, uh, where you're trying to minimize the amount of back and forth before a decision is made. Some places it seems like it's the opposite. They're finding they're trying unintentionally, maybe trying to maximize the amount of discussion. So how do you actually do that from a from an actual tactical perspective?

43:27Ali Ansari: I think there's two things. One is you have to reason through it with the with the team. Like when we have all hands, we remind folks of this. And there's always new of course, new people and all hands. So we gotta like kind of reason through from, from, from from first principles on why we set this the way it is And the reason is it goes back to, you know, this idea of like you, the length of the discussion being much longer. Will converge the same actions. You should, you should, you should stop that discussion sooner so that we can just act. And and the second component is that the velocity of your actions matter matters a lot because I look at it as like progress is sort of a, you know, sort of I'm simplifying it a little bit, but but it's sort of like the, the quality and accuracy of actions times the amount of actions. Right? That's sort of like the basic formula. The quality of actions come from these discussions, but also it just keeps improving over time as the company gets more, more sophisticated. It's a variable that you can that you can control with again these threads that, that, that that you open. But but you can't control it beyond like you can't like fundamentally change the quality for, for for all the actions that you have, you can uh, you can you can sort of iterate on them based on these discussions a little bit. But but you must. You must just act and and so for you to make progress, the variable that you can control even more so is the velocity part is the velocity of action. So so we just reason through it with the team. And then this makes sense. If everyone you know if anyone has any counters we sort of like discuss it. Um, and and yeah I think it's hard to sort of debate against it. And, and then the other thing is like we when, when folks join us, it's a bit of a shocking thing because like a lot of folks are, are, uh, they, they're used to meetings and they, they like, um, you know, they like discussing things and maybe like, hanging out a little bit in these meetings and and we don't we don't do that. We just, uh, and we just remind them of that. I think, like, whenever, whenever there's new people that join us and we realize like, uh, they're sort of they're sort of suggesting a meeting and a thread to, to many times, like sometimes a thread can't close because it's too complex and you got to do a meeting. But but if you default to like, let's do a meeting and discuss this live and let's do it next week or something. We call it out like you should. You should never. You should never suggest circling back next week like these are these are phrases that we avoid. And um, and I think if for new people like if you just if you just remind them a few times, you just build a feedback loop and and they, they, they end up doing the, doing the same thing.

45:55Julian Weisser: They get to see the results of what this culture drives and exactly motivates them and helps them sort of adapt to it.

46:01Ali Ansari: Exactly.

46:02Julian Weisser: That's really cool. What what are some things that if somebody joined the company, uh, you know, today, uh, and they were experiencing their first, maybe 30 days, things that would surprise them in a positive way. Uh, because I'm curious, you know, that that would ultimately be probably some interesting things for us to discuss.

46:20Ali Ansari: I think the first would be the we have a lot of really cool clients like, and we're very, you know, it's always an honor that we that we get to work with these clients. And not every not every new team member gets to get to see like, exactly all the clients because, you know, we have like, silos and so forth. But but but but once you get a good, good picture of like all the companies you work with, it's like, I mean, to me, it's like to me, it's also unbelievable that we get to work with all these, these companies. So that's always a bit of a surprising. It's a bit of a nice surprise where we're like especially a lot of like great researchers that end up working at some of these labs. They they realize that if they join micro1, they can they can work at they can basically work at multiple labs, you know, in sort of indirect ways. And so I think there's there's really nice nuances to that. Um,

47:08Ali Ansari: I think the other thing is like there's and this one might be, uh, you know, this one's more of a neutral thing, which is they realize how, how fast paced micro1 is. And of course, it is easy to say and everyone says it, but, but, but we people, people just often describe micro1 as it redefines their perception of fast pace. And this goes back to just this idea of acting and reducing and minimizing discussions. So that's that. That comes as a nice, you know, nice surprise. Um, and then I think the third thing is like, even though we're even though everything's crazy and stressful and so forth. Everyone's very supportive. And the thing that we emphasize. Like when somebody joins us as a new team member is like, we got to make sure the field they feel the microphone culture in terms of how how like difficult and hardcore it's going to be. They need to feel that right away because you can't get the wrong perception. So we make sure that like, you know, they feel a bit of pressure right away. And then and then the other thing is like we want to make sure they feel very well supported. Those those two things need to be true at the same time. So I think the kindness and the support that, that that team members see right away is, is a hopefully not a surprise. But but I think it's a little bit surprised because it's, it's a, it's like more than what you would expect.

48:23Julian Weisser: Now you're in a massive market. Uh, in fact, you're in many massive markets. Um, something that's really cool about micro1 is just how many different types of customers you serve. Um, I think that often case, uh, with startups where they have something that's working, they have lots of opportunities that come with that. Yeah. And there's some sometimes, uh, you know, a debate as to which opportunities to choose. Right. Uh, maybe just internally as the, as the, the founder. Um, then you start to have other teammates involved in making decisions as well. But I'm very curious to hear kind of what you've learned about, uh, about which opportunities to choose how to how to stage and sequence opportunities. And then, you know, maybe we could also talk about Spirit Airlines because Spirit Airlines is this is this crazy sort of opportunity that seemingly just kind of appeared. And it wasn't necessarily I can't imagine it was on your opportunity set. Um, and so maybe we could use that as sort of a case study as sort of an anomalous opportunity. And then we could also talk about, you know, the more sort of standard thing, which is, you know, you're building at a fairly broad, you have a fairly broad potential market to sell to. I'm curious if we could talk about that as well, but maybe the airlines as a case study?

49:36Ali Ansari: Yeah. Spirit is a is a is a definitely an outlier. So spirit is um, there's essentially.

49:46Ali Ansari: So let me go back to our actually so what we do at micron, which I realize we haven't talked about is we we build environments, right. We build environments for for labs to train their models. And

49:58Ali Ansari: what's included in these environments is, um, is is is seeding data, tooling and tasks. And the tasks are what agents or models are prompted to do. And then there's verifiers that determine whether they do a good job. So so there's a bunch of ingredients in an environment. And one of the things that matters a lot, which we sort of like look at it as like a bit of a bit of a scaling law. It's like, you know, you can sort of scale on this, which is, which is the realism of the environment. How realistic are the tasks, how realistic is the seeding data, the tooling, etc. and the way that you make environments more realistic is you is well, you need real data, right? And real data comes from companies that have that have operated in a production setting. And and so we have this function which we call company data partnerships, which is we call CDP, which partners with lots of companies. It's a bit of a crazy thing, actually. It's a we have about 150 companies that are signing up every single day right now to, uh, to partner with us. And it's the funnel is wild. Um, and we essentially partner with these companies, we license their data, we anonymize it, and then we use it for for training. We use it to seed the environments that we have. We build the verifiers on top of them, and then we train models, uh, with them. And the thing that matters a lot here is, again, the realistic nature of these environments. And company operational data allows us to make it very realistic. Spirit is just one another example that, uh, it's, you know, it's an outlier example. It's the one that makes the news and, uh, and so forth. But but it's just another company that in this case is very, very massive that, that, that we're buying, uh, that we're attempting to buy, um, to, to seed many different environments to, to, to help train models. Um, so that's kind of the tldr. And uh, there's a, there's a hearing coming up on the September 30th that that I think will, uh, will probably give the outcome of this, of this, of this bid. But but it might go a little longer as well. So we'll see.

51:54Julian Weisser: So okay, so that's, that's really interesting and sort of how it fits into the broader scheme of things. But I'm really curious uh, as we talk about sort of, you know, what's going on with, um, with the various types of customers you work with. Um, you know, and this is actually ties into something that you just announced, I think, yesterday, which is, you know, you're you're building out, um, you know, an office at a space in Malibu. Right. Um, so maybe you can talk about, uh, what's going on there? Uh, when to decide as a founder, when to decide to move into sort of adjacencies because robotics, I think, you know, maybe a year ago wasn't something that was immediately on the roadmap. So I'm curious how you think about that, how founders should think about that. When did when to do it. What's the right time?

52:38Ali Ansari: Yeah. Yeah. So I think what's important to, to us at micro1 is we need to have a very clear core focus. And that core focus is everything around the data stack. And what I mean by that is, you know, there's sort of three ingredients to improve models, right? There's data compute and algorithms. And and it's not just models. It's just improving intelligence systems is the way to think about it. It's agents, models, hardware embodiments and models. You know, all these different intelligence systems that can act in the world. So our entire focus is everything around the data stack, which is really the key ingredient that allows models and agents to to improve. So anything that that is around that we we can, we can we can expand into that. That's sort of the way we think about it. And, and so and like the sort of like abstract way to explain, you know, micro One which, which, which, which gives us which will, which will convey the, the way that we decide these, these, um, different businesses to get into is we're in the business of providing what we call units of intelligence improvement. So similar to how labs provide units of intelligence, they provide tokens, which is like just intelligence. We provide intelligence improvement units. So it's a very, very clear distinction. And

53:57Ali Ansari: today that comes in the form factor of data sets and environments that we're building. But the form factor will change over time. The form factor will will at some point probably be roll out to the service. Or it might be reward model service where we're moving up the stack. Right. It might be post training as a service. And and also like the areas, the sort of models and agents and intelligence systems that that require this improvement are also very different. It could be models, it could be agents within the enterprise. It could be a humanoid robots. It could be different hardware embodiments that look nothing like a humanoid. So as long as we have a core mission, which is we want to build predictable and abundant, we want to build a data engine that produces predictable and an abundant units of intelligence improvement. Anything that's related to that core mission that we think we need to expand to, we can expand to. And LM of course, serving the LM companies is a clear, clear area to be in serving enterprises and is another clear area to be in because if enterprises want to own their own intelligence, they need to build their eval stack. Robotics is if you want to build models that can act in the physical world, you need to go into robotics where there's a ton of value to be had there. So so all of these, these areas that we're in, they are under the umbrella of, of of a data engine that can that can produce these intelligence improvement units. That's sort of how we how we think about it.

55:18Julian Weisser: It's really cool. And specifically Malibu. Why Malibu? What's the what's the what's the idea there? Right now, we're filming at your space in San Francisco. You have most of the team is distributed. Yeah. So you've got some incredible folks who I. We all got to meet as we were setting up. Uh, but most people are, you know, not in SF. So what about Malibu?

55:41Ali Ansari: I think that the the right question is why not Malibu and why?

55:45Julian Weisser: Well, sure, I agree.

55:47Ali Ansari: I know the funny one liner is like, we you know who doesn't want to train robots at the beach, right? That's a that's sort of the the funny way to think about it. But, but but there's a couple of reasons. One is getting researchers is very difficult. And in robotics it's maybe even more difficult. And and we need a very unique angle to, to attract researchers. And this one is like is very unique. You gotta you join us and you can you can train robots at the beach and it's it'll be a good time. That's the first thing. The second thing is our team is distributed anyways. And, uh, and we're not like, we're not set on any location. Of course, our main HQ is NSF. We have good, good amount of folks here, but but we can be anywhere so and and and one of our, a couple of our key folks in robotics teams actually are actually in LA. So so why not why not have it in LA. And then the third component is um, LA is known for hardware broadly, uh, of course, like the humanoid robotics companies are not necessarily in LA, but but there's a lot of other hardware companies that are we believe in a world where there's going to be a ton of policy models that are built for many different hardware embodiments, not just humanoids. And so I think the hardware companies that are in LA are going to be very important for the work that we do. So we want to be closer to them as well. So that's another, you know, another reason. So and it's just like it's just cool. Like you know, it's going to be I think it's going to be a lot of fun to, to, to uh, train robots at the, at the beach.

57:09Julian Weisser: So before we ask sort of the, the customary closing questions, I had one one more question for you, uh, which is about sort of values, uh, because I think that, you know, one thing I read you talk about, uh, is about sort of, you know, the state of the world and sort of, um, like working with the US and the allies or allies and some people potentially selling data to, you know, adversarial countries or and I'm curious, you know, what your perspective is on this kind of how you're trying to build the, the company, um, as a way to, you know, help, help the US and help our allies grow. Yeah.

57:47Ali Ansari: I, I really appreciate this question. The approach we take at micro1 generally is

57:54Ali Ansari: everything we do, we we take a very long term outlook and a very principled, uh, approach. And

58:02Ali Ansari: a lot of times the principled approaches that we take, they do not result in hundreds of millions and run rate increases immediately. And in other words, they don't result in us necessarily being the number one player in terms of revenue in the short and medium term. But what they do in the long term that, that, that, that, that, that's, that's the that's sort of the high level, uh, reason for these decisions. And the decision of working with only us and allies is, is, is simply a good example of, of, uh, of this, this, this idea of like being very principled in the way that we in the way that we work, even if it means short term or medium term revenue losses. And we think winning, uh, the iris is a single most important thing that that, that any country can do. And, you know, I, uh, I love the US. I'm very grateful for the US. I, I, uh, when I moved from Iran, when my family won the green card lottery, um, I, we were we were like, so excited to move and, and the opportunities of this country gave me and my family is, uh, is just like, I can't put into words how grateful I am for it, and the absolute least I can do is is is help this country win what I think is the most important race to win, which is AI. Um, and, and I think winning the air race and of course, you know, biased when I say this, but but I think data is the most important part of winning the race. And, and if you work with US labs and you know exactly the type of data that works well, the taxonomies, the domains, the data sets overall that work well, and you build large scale off the shelf data sets and you go and you sell them to adversarial nations. I think you're doing a huge disservice to to this country. And we are not in the business of, of of doing that. And and again, it hurts us with potentially nine figure run rate. Right. And revenue. But but but that doesn't matter because these are these are the decisions that allow us to win long term. And we're very, very proud of these. Uh, very proud of these, um, very proud of these decisions. And I think, you know, I've said this before, and I'm more than happy to say it now. Data companies that work with adversarial nations,

60:26Ali Ansari: they should just be ashamed. I think it's a very shameful act. And I think they're they're helping with, with, uh,

60:34Ali Ansari: they're essentially helping with the, the distilling of, of us models is the reality of it. And, uh, and I and I hope this changes and, and I and I hope that there's, uh, uh, there's some, there's some government agencies that look into to building frameworks around regulating this because if you again, exporting data is, is at a, at a large scale in a very structured way. Because if you think about what an environment is, it's it's just structured intelligence at a large scale is what it is like. That's literally what it is. And and especially if you include company operational data, it's structured intelligence at a large scale. And. Company data from US companies. If you if you package that into environment and you sell it to adversarial nations, you're exporting American intelligence. And that is like that's absurd to me that, that, that, that, that companies are able to do that. Um, so anyways, a bit of a bit of a speech, but but I think again, like, I'm grateful for the US and I and I, uh, and I just want to do what's, what's what's right for this country and um, and, uh, and that's, that's, that's, that's our reasoning for this approach.

61:40Julian Weisser: It's a really, uh, it's a really stirring way that you put it. I don't think I've ever heard it put quite that way. The idea that this is like the collective intelligence of America that is being sold literally wholesale to our adversaries, it's not just, you know, in a lot of these folks, a lot of the intelligence comes from the educational systems in this country, which is our taxpayer money. It's like all of these all of these things kind of add up to be do something that, you know, is it you you want to see this country when? Um, and and it's it's a shame that some people are, you know, uh, maybe, maybe taking short term, uh, sort of short term decisions, uh, maybe, uh, very real consequences to us as a nation.

62:24Ali Ansari: 100%.

62:25Julian Weisser: Well, um, this has been amazing. We have we have two sort of final closing questions. We always go with. The first one is the case against solo founding, like the bear case. Why? Maybe somebody should maybe consider not being a solo founder. Now, I know for you, uh, it wasn't even ever a consideration, which I think is great. In fact, some of the best solo founders we met, they come from different places where they didn't have sort of the doctrines of Silicon Valley, which I think has helped many companies get started, but probably caused many great potential companies to fail, um, through, through these various beliefs. I'm curious sort of what your what your take is though, on what reasons people should be cautious about being a solo founder Before we get into the preview, the final question, which is the case for solo founding?

63:10Ali Ansari: So I think the bear case is, uh,

63:15Ali Ansari: it's definitely a lot more lonely. Um, but but but to be honest, I don't think for too long. Like, if you if you, if you build a good founding team, it's not actually going to be that lonely for that long. So it's a little bit of a bear case, uh, in terms of being being more lonely, I think

63:33Ali Ansari: the other thing is, um,

63:36Ali Ansari: if you're building a really, like, difficult company, like, if you're like, you know, like hard mode hardware or maybe you're building some, uh, I dunno, something that, like, really requires a ton of specialty expertise that, that all needs to be deeply committed to that thing for it to work from the beginning. Then then there's there's potentially a case for soul for, for co-founding with, with a couple of folks that have these different expertise that are complementary, but I think this is more of like a once you build a couple of companies that work and you go in like hard mode for, for your, for your like, ultimate company or whatever. At that point, like maybe it's good to, to explore. Yeah, I think those are those are the two things that come to mind which like, are not strong arguments, to be honest.

64:21Julian Weisser: Yeah. Well, I mean, especially because once you start to have a business with customers and you start to really bring on the team, you said the loneliness starts to go away really quickly. If you actually have a true team that you're treating like a founding team, if you don't, then that's kind of your fault. I think at that point, if you're if you're still lonely after that. And the other thing about sort of the deep tech and sort of hard, hard mode startups, I've been surprised by just how many companies out of El Segundo, for instance, are solo founded. Yeah. Um, these are these are companies that you would kind of think might require, uh, co-founders. But I think perhaps this is what I've heard from some is that some of the people that you want to hire, uh, early on, they wouldn't want to be co-founders. They have different incentive mechanisms in terms of what they care about and sort of when they want to join. So that's that's been interesting to see. I this has been amazing. Thank you so much for doing this. Let's let's close it out. What's the case for solo founding?

65:12Ali Ansari: Yeah there's a lot there's a lot I think the first is you can make a lot, a lot faster decisions. Um, uh, you know, there's, there's sort of one person, like, clearly leading and I think there's, there's a lot of value to that. There's no arbitrary constraints of like, let me do co-founder matching. And, you know, you don't have to wait for that. You can just start start the thing. You get to control, uh, you get to control the board more which which I think is, is important. And this is something that, you know, we can get into in maybe the second episode, if we do one where it's like we, you know, I care a lot about and I look at it as like company control board. It's not like, oh, I, you know, I control it's more of like company controls board. And and again, it goes back to like we are so in the weeds of these like really difficult decisions that make this outlier outcome. We must be the ones that make Stem. We can't have, you know folks that, that, that, that are less in the weeds. Um. Um. Uh, be too involved. So, um, so, yeah, I think that's another thing.

66:08Ali Ansari: The third is, um, I think if you build, if you sort of build something, if you get something to a reasonable stage, uh, for, for, you know, for a few months on your, on your own, you actually are. There's, there's arguments you made that you can get there faster and, and you can also attract the founding team that you need potentially faster as well, which is what really allows you to kind of kick things off and have things kind of take off. So I think that's, that's that's the other thing that I would refer to. Um, and then I mean, the last part is like you get to own more of the more of the company. And yeah, this was a selfish one, but like, it's it's just the reality of it.

66:50Julian Weisser: This has been amazing. Thank you so much for doing this. Uh, I'm excited to, uh, do the next one where we talk about governance and all sorts of, miles. Uh, yeah. It's gonna be great. Thank you. Thank you so much. And, uh, excited for people to listen to this.

67:04Ali Ansari: Yeah. Thank you very much. It was awesome. Thanks for having me.

67:07Julian Weisser: If you enjoyed this conversation, we would love it. If you can share solo founders with everyone you know, particularly people who are considering starting companies solo or who are already doing it. You know, the entire goal of what we do is to normalize solo founder and encourage more people to solo founded instead of end up with co-founders of convenience, if you'd like, and we'd really appreciate it. You can go leave a review on Apple. You can go give us a thumbs up. You can subscribe on the YouTube channel. Anything helps. And of course, if you are starting a company or considering starting a company, you should go look at the Solo Founders Program. We work together with you in San Francisco alongside of a bunch of other solo founders. The idea is it's much better to be solo together than to be solo alone. I'll spend a lot of time helping you, and we also invest $100,000 in your company. So if you're excited to learn more about that, go check out solofounders.com/program.

If you're already solo, you're early.

Whether you're solo founding now or still exploring, stay in the loop.

© 2026 Solo Founders. All rights reserved.

Privacy PolicyTermsSan Francisco, California