Pieter Levels: Programming, Viral AI Startups, and Digital Nomad Life | Lex Fridman Podcast #440

Pieter Levels 0:00 So I was trying to figure out how to do photo-realistic AI photos, and it was... Stable diffusion by itself is not doing that well. Like, the faces look all mangled.

Lex Fridman 0:06 Yep.

Pieter Levels 0:07 Um, and it doesn't have enough resolution or something to, to, to do that well. So... But I started seeing these, these, m- um, base models, these fine-tuning models, and people would train on porn. And I would try them, and they would be very photo-realistic. They would have bodies that actually made sense, like body o- uh, a- a- anatomy. Um, but if you look at the photo-realistic models that people use now, still, there's still a core of porn there, like, of naked people.

Lex Fridman 0:31 Yeah.

Pieter Levels 0:31 So I need to prompt out the nak-... And everyone needs to do this with AI startups with imaging. You need to prompt out the naked stuff.

Lex Fridman 0:36 Do you have to keep reminding the model you need to put-

Pieter Levels 0:38 It's not-

Lex Fridman 0:38 ... clothes on the thing?

Pieter Levels 0:39 Yeah, "Don't put naked 'cause it's very risky." I have Google Vision-

Lex Fridman 0:42 Yeah.

Pieter Levels 0:42 ... that checks every photo before it's shown to the user to, like, check for-

Lex Fridman 0:45 Like a nipple detector?

Pieter Levels 0:46 ... NSFW. Yes.

Lex Fridman 0:47 Oh (laughs) NSFW de-, detector.

Pieter Levels 0:48 Because you get the journalists get very angry.

Lex Fridman 0:52 The following is a conversation with Pieter Levels, also known on X as LevelsIO. He is a self-taught developer and entrepreneur who designed, programmed, shipped, and ran over 40 startups, many of which are hugely successful. In most cases, he did it all by himself while living the digital nomad life in over 40 countries and over 150 cities, programming on a laptop while chilling on a couch, using vanilla HTML, jQuery, PHP, and SQLite. He builds and ships quickly and improves on the fly, all in the open, documenting his work, both his successes and failures, with a raw honesty of a true indie hacker. Pieter is an inspiration to a huge number of developers and entrepreneurs who love creating cool things in the world that are hopefully useful for people. This was an honor and a pleasure for me. This is the Lex Fridman Podcast. To support it, please check out our sponsors in the description. And now, dear friends, here's Pieter Levels.

Lex Fridman 2:03 You've launched a lot of companies and built a lot of products. As you say, most failed, but some succeeded. What's your philosophy behind building the startups that you did?

Pieter Levels 2:14 I think my philosophy is very different than most people in startups, 'cause most people in startups, they, they build a company, and they raise money, right? And they hire people, and then they build a product, and they find something that makes money. And I don't really raise money. I don't use VC funding. I do everything myself. I'm a designer. I'm the developer. I, I make everything. I make the logo. So for me, I'm much more scrappy. And-

Lex Fridman 2:36 (laughs)

Pieter Levels 2:36 And because I don't have funding, like, I need to, I need to go fast. I need to, uh, make things fast, uh, to see if an idea works, right? I have an idea in my mind, and I build it, build, like, a micro, mini startup, um, and I launch it very quickly, like within, you know, two weeks or something of building it. And I check if there's demand and if people actually sign up. And not just sign up, but if people actually pay money, right? Like, they need to take out their credit cards, pay me money, and then I can see if the idea is validated. And most ideas don't work, like, as you say, most fail.

Lex Fridman 3:05 So there's this rapid iterative phase where you just build a prototype that works, launch it-

Pieter Levels 3:11 Yeah.

Lex Fridman 3:12 ... see if people like it, improving it really, really quickly to see if people like it a little bit more, enough to pay and all that. That, that whole rapid process is how you think of-

Pieter Levels 3:21 Yeah. I think it's, it's like, it's very rapid. And it's like, um, if I compare it to, for example, Google, you know, like, or big tech companies. Especially Google right now is kind of struggling. Like, they made, like, transformers. They made all the... invented all the AI stuff years ago, and they never really shipped L- like, they could have shipped ChatGPT, for example, I think I heard in 2019. And they never shipped it because they were so stuck in bureaucracy. But they had everything. They had the data, they had the tech, they had the engineers, and they could, didn't do it. Um, and it's because these big organizations, it's, it's, it can make you very slow. So being alone by myself on my laptop, like, you know, in my underwear in a hotel room or something, I can ship very fast. And I don't need to, like, I don't need to ask, like, legal for, like, "Oh, can you vouch for this?" You know? I can just go and ship.

Lex Fridman 4:02 Do you always code in your underwear? The- your profile picture, you're, like, slouching-

Pieter Levels 4:07 Yeah.

Lex Fridman 4:07 ... on a couch in your underwear, chilling on a laptop.

Pieter Levels 4:09 No, no, but it's... I w- do wear, like, shorts a lot. And I, I usually just wear shorts and no T-shirt 'cause I'm always too hot. Like, I'm always overheating.

Lex Fridman 4:16 Mm-hmm. Thank you for showing up not just in your underwear, but-

Pieter Levels 4:18 Yeah.

Lex Fridman 4:19 ... wearing shorts.

Pieter Levels 4:19 I know. You know, I'm still wearing this for you, but-

Lex Fridman 4:21 Thank you.

Pieter Levels 4:21 Um-

Lex Fridman 4:22 Thank you for dressing up.

Pieter Levels 4:23 (laughs) I think it's c- 'cause I, since I go to the gym, I'm always too hot.

Lex Fridman 4:26 What's your favorite exercise in the gym?

Pieter Levels 4:28 Man, overhead press.

Lex Fridman 4:29 Overhead press, like shoulder press?

Pieter Levels 4:30 Yeah.

Lex Fridman 4:31 Okay.

Pieter Levels 4:31 But it feels good 'cause you're doing, like... You do, you win. Uh, 'cause when you, like, I do-

Lex Fridman 4:35 (laughs)

Pieter Levels 4:35 What is it? I do 60 kilos, so it's like-

Lex Fridman 4:37 Yeah.

Pieter Levels 4:37 ... 120 pounds or something like this. It's-

Lex Fridman 4:38 Yeah.

Pieter Levels 4:38 It's my only thing I can do well, you know, in the gym. And you stand like this, and you're like, "I did it," you know? Like a winner pose.

Lex Fridman 4:45 It's a victory thing. Yeah.

Pieter Levels 4:45 A victory pose. Uh, I do bench press, squats, dead lifts. (laughs)

Lex Fridman 4:49 Hence the, uh, the mug.

Pieter Levels 4:50 Yes.

Lex Fridman 4:50 Talking to my therapist.

Pieter Levels 4:52 Yeah.

Lex Fridman 4:52 It's a dead lift.

Pieter Levels 4:53 Yeah. Because it acts like therapy for me, you know?

Lex Fridman 4:55 Yeah, yeah. It is.

Pieter Levels 4:55 Which is controversial to say. Like, if I say this on Twitter, people get angry.

Lex Fridman 4:59 Physical hardship is a kind of therapy.

Pieter Levels 5:01 Yeah.

Lex Fridman 5:01 I just re-watched, uh, Happy People: A Year in the Taiga, that, uh, Werner Herzog film where they document people that are doing trapping. They're essentially just working for survival in the wilderness year round.

Pieter Levels 5:15 Yeah.

Lex Fridman 5:15 And there's a deep happiness to their way of life-

Pieter Levels 5:18 Yeah.

Lex Fridman 5:18 ... because they're so busy in it, in nature.

Pieter Levels 5:22 Yeah, 100%.

Lex Fridman 5:22 Like, there's something about that physical-

Pieter Levels 5:24 Physical, yeah.

Lex Fridman 5:25 ... toil.

Pieter Levels 5:25 Yeah. My dad taught me that. My dad always d- does, like, uh, construction in the house. Like, he's always re-, renovating the house. He breaks through one room, and then he goes to the next room. And he's just going a- in a circle around the house for, like, the last 40 years.

Lex Fridman 5:37 (laughs)

Pieter Levels 5:37 So, but so he's always doing construction in the house, and it's his hobby. And he, like, he taught me when I'm, when I'm depressed or something. Um, he says, like, "Get a big, like..." What do you call it? Like, a big, uh, mountain of sand or something-

Lex Fridman 5:49 Mm-hmm.

Pieter Levels 5:49 ... from construction and just get a shovel. And, uh, bring it to the other side. And just, you know, do, like, physical labor. Do, like, s- hard work-

Lex Fridman 5:57 Mm-hmm.

Pieter Levels 5:57 ... and do something. Like, get, set a goal, do something. And I, I kind of did that with startups too.

Lex Fridman 6:02 Yeah, construction is not about the destination, man. It's about the journey.

Pieter Levels 6:06 (laughs) Yeah.

Lex Fridman 6:07 Yeah. Sometimes I wonder, people who are always remodeling their house, is it really about the remodeling or is it about-

Pieter Levels 6:11 No, no, it's not.

Lex Fridman 6:12 ... is it about the project?

Pieter Levels 6:13 It's about the journey.

Lex Fridman 6:13 The puzzle of it.

Pieter Levels 6:14 No, he doesn't care about the result. Well, he shows me. He's like, "It's amazing." I'm like, "Yeah, it's amazing." But, um, then he wants to go to the next room, you know? But I think it's very metaphorical for work, 'cause I also, I never stop working. I go to the next website-

Lex Fridman 6:27 Mm-hmm.

Pieter Levels 6:28 ... or I make a new one, right, or I make a new startup. So I'm always like, like it gives you something to wake up in the morning and like, you know, have coffee and then, uh, kiss your girlfriend and then you have like a goal. Now today I'm gonna fix this feature, today I'm gonna fix this bug or something. I'm gonna do something. You have something to wake up to, you know? And I think, um, maybe especially as a man, but also women, but you need, you need a hard work, you know? You need like an endeavor thing.

Lex Fridman 6:52 How much of the building that you do is about money? How much is it about just a deep internal happiness?

Pieter Levels 6:59 It's really about fun 'cause I was, 'cause I was doing it when I didn't make money, right? That's the point. So I was always coding, I was always, I was making music, I made electronic music, uh, drum and bass music like 20 years ago. And I was always making stuff. So I think, uh, creative expression is like a meaningful work that's so important. It's so fun. It's so fun to have like a daily challenge where you try figure stuff out.

Lex Fridman 7:20 But the interesting thing is you built a lot of successful products and you never really wanted to take it to that level where you scale real big-

Pieter Levels 7:29 Yeah.

Lex Fridman 7:29 ... and sell it to a company, something like this.

Pieter Levels 7:32 Yeah. The problem is I don't dictate that, right? Like if more people start using it, if millions of people suddenly starts using it and it becomes big, um, I'm not gonna say, "Oh, stop signing up to my website and pay me money." But I never raised, uh, funding for it. And I think 'cause I don't like the, the stressful life that comes with it. Like I have a lot of, um, founder friends and they tell me secretly, like with hundreds of millions of dollars in funding and stuff, and they a- they tell me like, "Next time, if I'm gonna do it, I'm gonna do it like you because it's more, it's more fun, it's more indie, it's more chill, it's more creative." They don't like this, they don't like to be manager, right? You become like a CEO, you become a manager and, um, I think a lot of people that start startups, when they become a CEO, they don't like that job actually, but they can't really exit it, you know?

Lex Fridman 8:16 Mm-hmm.

Pieter Levels 8:16 But they like to do the groundwork, the coding. So I think that keeps you happy, like doing something creative.

Lex Fridman 8:24 Yeah. It was interesting how people are pulled towards that, to scale, to go really big.

Pieter Levels 8:29 Yeah.

Lex Fridman 8:29 And you don't have that honest reflection with yourself, like what actually makes you happy. 'Cause for a lot of great engineers, what makes them happy is the building, the, the "individual contributor".

Pieter Levels 8:41 Yeah.

Lex Fridman 8:41 Like where you're actually still coding or-

Pieter Levels 8:43 Yeah.

Lex Fridman 8:43 ... you're actually still building. And they let go of that and then-

Pieter Levels 8:46 Yeah.

Lex Fridman 8:46 ... they become unhappy. Um, but some of that is the sacrifice needed to have a impact at scale if you truly believe in a thing you're doing.

Pieter Levels 8:55 Well, like, look at Elon. He's doing things million times bigger than me, right? And, um, would I want to do that? I don't know. You cannot really choose these things, right? But I really respect that. I think Elon's very different from VC founders, right? VC start, it's like software, there's a lot of bullshit in his world, I think. There's a lot of like dodgy finance stuff happening there, I think. Um, and I never have like concrete evidence about it, but your gut tells you something's going on with like companies getting sold to-

Lex Fridman 9:20 (laughs) Yeah.

Pieter Levels 9:21 ... friends and VCs and then they do reciprocity and there's shady financial dealings. With Elon, there's not. He's just raising money from investors and he's actually building stuff. He needs the money to build stuff, you know, hard- hardware stuff. Um, and that I really respect.

Lex Fridman 9:34 (inhales) You said that there's been a few low points in your life, that you've been depressed, and the building is one of the ways you, you get out of that. But c- can you talk to that? Can you take me to that place, to that time when you were in a, at a low point?

Pieter Levels 9:47 So I was in Holland and I graduated university and I didn't want to like get a normal job. And I was making some money with YouTube 'cause I had this music career and I uploaded my music to YouTube. And YouTube started paying me like, uh, uh, with AdSense, like $2,000 a month, $3,000 a month. And all my friends got like normal jobs and we stopped hanging out 'cause people would... Like in university, you hang out, you know, you chill with each o- at ch- at each other's houses. Uh, you go party. But if and when people get jobs, they only party like in the weekend and they don't hang anymore in the week 'cause you need to be at the office. And I was like, "This is not for me. I want to do something else." And I was starting getting this like, I think it's like Saturn return this, you know, when you're t- turn 27. It's like some concept where Saturn returns to the same place in the orbit that it was when you're born. Man, it's like-

Lex Fridman 10:28 I'm learning so many things.

Pieter Levels 10:29 ... it's some astrology thing, you know?

Lex Fridman 10:31 So many truly special artists died when they were 27.

Pieter Levels 10:34 Exactly. Something with 27, man. And it was for me. Like I started going crazy.

Lex Fridman 10:38 Yeah.

Pieter Levels 10:38 Because I didn't really see like my future in Holland, buying a house, going living in the suburbs and stuff. So I flew out. I went to Asia, started digital nomading and did that for a year. And then that made me feel even worse, you know, 'cause I was like alone, um, in hotel rooms, like looking at the ceiling, like, "What am I doing with my life?" Like, this is a... Like I was working on startups and stuff and YouTube, but it's like, "What is the future here?" You know? Like, uh, is this, is this something... While my friends in Holland were doing really well and with a normal life, you know? Um, so I was getting re- very depressed and like, I'm like a outcast, you know. I'm... My money was, was shrinking. I wasn't making money anymore a lot, was making $500 a month or something. And I was, you know, looking at the ceiling thinking like, "Now I'm like 27. I'm a loser." And that's the moment when I started building like startups. And it was because my dad said like, "If you're depressed, you need to, you know, get sand, get a shovel, start shoveling, doing something. You can't just sit still." Which is kind of like a interesting way to deal with depression, you know. Like it's not like, "Oh, let's talk about it." It's more like, "Let's go do something."

Lex Fridman 11:38 Mm-hmm.

Pieter Levels 11:39 And, um, and I started doing a project called 12 Startups in 12 Months where every month I would make something, like a project, and I would launch it with Stripe so people could pay for it.

Lex Fridman 11:49 So the basic format is try to build a thing, put it online and put Stripe to where you can pay money for it.

Pieter Levels 11:55 Yeah. Add a Stripe check. I'm not sponsored by Stripe, but add a Stripe checkout button.

Lex Fridman 11:58 Is that still like the easiest way to just like pay for stuff, Stripe?

Pieter Levels 12:02 100%. Like I think so, yeah.

Lex Fridman 12:03 It's a cool company. They just made it so easy. You can just click and-

Pieter Levels 12:06 Yeah.... and they're really nice. Like the CEO, Patrick, is really nice.

Lex Fridman 12:09 Behind the scenes, it must be difficult to, like, actually make that happen. 'Cause that used to be a huge problem, like just-

Pieter Levels 12:15 Merchant?

Lex Fridman 12:15 ... just, just adding a thing, a button, where you can, like, pay for a thing.

Pieter Levels 12:20 Dude.

Lex Fridman 12:20 It's-

Pieter Levels 12:20 Dude, I know this because when I was a, when I was-

Lex Fridman 12:22 Trustworthy.

Pieter Levels 12:23 ... nine years old, I was making websites also. And I tried to open a merchant account. That was like before Stripe, you would have like, um, I think it was called WorldPay. So I had to, like, fill out all these forms, and then I had to fax them to America from Holland, with my dad's fax. Um, and my dad had to s- it was in my dad's name, and he had to sign for this. And he started reading these terms and conditions, which was like, he's liable for like 100 million in damages.

Lex Fridman 12:47 Mm-hmm.

Pieter Levels 12:48 And he's like, "I don't (laughs) want to sign this." I'm like, "Dad, come on. I need a merchant account. I need to make money on the internet, you know?"

Lex Fridman 12:52 Yeah. (laughs)

Pieter Levels 12:53 And he signed it, and we sent it, we faxed it to America, and I had a merchant account. But then never, nobody paid for anything, so that was the problem, you know?

Lex Fridman 12:59 Yeah.

Pieter Levels 12:59 But it's much easier now. You can sign up, you add some codes, and, um,

Pieter Levels 13:03 yeah.

Lex Fridman 13:03 So 12 startups in 12 months?

Pieter Levels 13:05 Yeah.

Lex Fridman 13:06 So what, (laughs) how do you, what, startup number one, what, what was that, what, like what, what were you feeling? What were you sitting behind the computer, like how much did you actually know about building stuff at that point?

Pieter Levels 13:18 Well, I could, I could code a little bit 'cause I did the YouTube channel. And I made a website for, I would make websites for like the YouTube channel, it was called PANDAMixShow, and it was like these electronic music mixes, like dubstep or drum and bass or techno or house.

Lex Fridman 13:29 I saw one of them had like Flash. Were you using Flash?

Pieter Levels 13:32 Yeah, my album, my CD album was using Flash, yeah, yeah.

Lex Fridman 13:35 Yeah.

Pieter Levels 13:35 I sold my CD, yeah.

Lex Fridman 13:36 'Cause Flash was a-

Pieter Levels 13:37 Dude, Flash was cool.

Lex Fridman 13:38 ... software. This is like the-

Pieter Levels 13:39 Yeah.

Lex Fridman 13:39 ... the break that-

Pieter Levels 13:40 Like Grandpa, you know? But Flash was cool.

Lex Fridman 13:42 Yeah. And there was, uh, what's it called? Oh boy, I should remember this, ActionScript. This is some kind of programming language.

Pieter Levels 13:47 ActionScript, yeah, yeah, ActionScript.

Lex Fridman 13:48 Oh, yeah.

Pieter Levels 13:48 It was in Flash. Back then, that was the JavaScript, you know?

Lex Fridman 13:50 The JavaScript, yeah.

Pieter Levels 13:52 Yeah.

Lex Fridman 13:52 And I, I thought that's gonna, that's supposed to be the dynamic thing that takes over the internet.

Pieter Levels 13:56 Yeah.

Lex Fridman 13:56 I invested so many hours in learning that.

Pieter Levels 13:57 And Steve Jobs killed it.

Lex Fridman 13:58 Steve Jobs killed it.

Pieter Levels 13:59 Steve Jobs said, "Flash sucks. Stop using it." And everyone's like, "Okay."

Lex Fridman 14:02 That guy was right though, right?

Pieter Levels 14:04 Yeah. I, I don't know. Yeah. Well, it was, it was a closed platform, I think. And it-

Lex Fridman 14:08 Closed?

Pieter Levels 14:08 But it's ironic 'cause Apple, you know, they're not very open.

Lex Fridman 14:11 Right.

Pieter Levels 14:11 But back then, Steve want- was like, "This is closed. We should not use it." And it has security problems, I think. Which sounded like a cop-out, like he just wanted to say that to make it look kinda bad. Um, but Flash was cool. Yeah.

Lex Fridman 14:22 Yeah. It was cool for a time.

Pieter Levels 14:23 Yeah.

Lex Fridman 14:24 It... Listen, animated GIFs were cool for a time too.

Pieter Levels 14:27 Yeah.

Lex Fridman 14:27 They came back in a different way.

Pieter Levels 14:29 I still use... Yeah.

Lex Fridman 14:29 As a meme though.

Pieter Levels 14:30 Yeah.

Lex Fridman 14:30 I mean, like, I, I even remember when GIFs were actually cool, not ironically cool.

Pieter Levels 14:36 Yeah.

Lex Fridman 14:36 Like, there's like (laughs) on the internet, you would have like a dancing rabbit or something like this.

Pieter Levels 14:41 Yeah.

Lex Fridman 14:41 And that was really exciting.

Pieter Levels 14:42 No, you had like the, you know, Lex homepage.

Lex Fridman 14:44 Yeah.

Pieter Levels 14:44 You had all these... Everything was centered.

Lex Fridman 14:46 Yeah.

Pieter Levels 14:46 And you had like Peter's homepage, and then the under construction-

Lex Fridman 14:50 Yeah.

Pieter Levels 14:50 ... GIF, which was like a guy like-

Lex Fridman 14:52 (laughs)

Pieter Levels 14:52 ... with a helmet. And the lights. It was amazing.

Lex Fridman 14:54 And the banners, uh, that's how, before like Google AdSense, you would have like banners for advertisements.

Pieter Levels 15:00 It was amazing, yeah.

Lex Fridman 15:01 And, and a lot of links to porn, I think.

Pieter Levels 15:04 Yeah.

Lex Fridman 15:04 Or

Narrator 15:05 porn-type things.

Pieter Levels 15:05 I think that was where the merchant accounts, people would use for. People would make money, a lot of... The only money made on the internet then was like porn.

Lex Fridman 15:10 Yeah.

Pieter Levels 15:10 Or a lot of it. Yeah.

Lex Fridman 15:12 Yeah. It was a, it was a dark place. It's still a dark place.

Pieter Levels 15:15 Yeah.

Lex Fridman 15:15 And, but there's beauty in the darkness. Anyway, so you were, uh, you did some basic HTML.

Pieter Levels 15:20 Yeah. Yeah, but I had to learn the actual, like, coding. So I, uh, this was good. It was a good, good idea to, like, every month launch a startup, so I could, uh, learn to code, learn basic stuff. And, but it was still very scrappy 'cause I didn't have time to, which was on purpose, like, I didn't have time to spend a lot of, um... I had a month to do something, so I couldn't spend more than a month, and I was pretty strict about that. Um, and I published it as a blog post, so people... I think I put it on Hacker News, and people would check, like, kinda like, "Oh, did you actually," you know? I felt like accountability 'cause I put it public that I actually had to do it.

Lex Fridman 15:50 Do you remember the first one you did?

Pieter Levels 15:52 I think it was Play My Inbox, 'cause back then, my friends, we would send, we would send like cool mu- it was before Spotify, I think. We would send like, uh, 2013, we would send music to each other, like YouTube links.

Lex Fridman 16:03 Mm-hmm.

Pieter Levels 16:04 Uh, like, "This is a cool song. This is a cool song." And there was these, these giant email threads on Gmail. And they were like un-navigable, so I made a app that would log into your Gmail, get the emails and find the ones with YouTube links, and then make like, kinda like a gallery of your, your songs. Like, essentially Spotify.

Lex Fridman 16:20 Mm-hmm.

Pieter Levels 16:20 And my friends loved it.

Lex Fridman 16:21 Was it scraping it? Like, what was the API?

Pieter Levels 16:23 No, it uses like P-P-O-P, like pop, or IMAP, you know? It would actually check your email. So it had like privacy concerns 'cause it would get all your emails to find YouTube links.

Lex Fridman 16:31 (laughs)

Pieter Levels 16:31 But then I would, I wouldn't save anything. Um, but that was fun. That was like... And that, that first product already would get, like, press. Like it went on, I think, like, um, some tech media and stuff. And I was like, "That's cool." Like, it didn't make money. There was no payment button, but it was, uh, it was actually, people were using it. I think tens of thousands of peoples used it.

Lex Fridman 16:51 That's a great idea. I wonder why, like, why, why don't we have that? Why don't we have things that access Gmail and extract some useful aggregate information?

Pieter Levels 17:01 Yeah. Yeah, you could tell Gmail, like, "Don't give me all the emails. Just give me the ones with YouTube links," you know, or something like that. Yeah.

Lex Fridman 17:07 I mean, there is a whole ecosystem of, like, apps you can build on top of the Google.

Pieter Levels 17:10 Yeah.

Lex Fridman 17:11 But people don't really buil-

Pieter Levels 17:12 Never do this. Like I've never seen them.

Lex Fridman 17:13 They build... I've seen a few, like Boomerang.

Pieter Levels 17:16 Yeah.

Lex Fridman 17:16 There's a few apps that are, like, good, but just-

Pieter Levels 17:19 Yeah.

Lex Fridman 17:19 I wonder what... Maybe it's not easy to make money?

Pieter Levels 17:22 I think it's hard to get people to pay for these, like, extensions and plugins, you know?

Lex Fridman 17:25 Yeah.

Pieter Levels 17:25 'Cause it's not like a real app, so it's not like... People don't value it. People value, "Oh, this..." And a plugin should be free, you know? When I want to use a plugin in Google Sheets or something, I'm not gonna pay for it. Like, it should be free. Which is... But if you go through a website, and you actually, "Okay, I need this product. I'm gonna pay for this 'cause it's a real product." So even though it's the same code in the back, it's a plugin, you know?

Lex Fridman 17:44 Yeah. I mean, you could do it through, like, extensions, like Chrome extensions, through, from the browser side.

Pieter Levels 17:49 Yeah, but who pays for Chrome extensions, right? Like barely anybody.

Lex Fridman 17:51 Nobody.

Pieter Levels 17:51 So-

Lex Fridman 17:51 Nobody.

Pieter Levels 17:52 ... uh, that's not a good place to make money probably.

Lex Fridman 17:54 Yeah, that sucks.

Pieter Levels 17:55 Like Chrome extensions should be a extension for your startup, you know? You have a product.

Lex Fridman 17:59 Yeah.

Pieter Levels 17:59 Oh, we also have a Chrome extension, you know?

Lex Fridman 18:02 I wish the Chrome extension would be the product. I wish Chrome would support that.

Pieter Levels 18:06 Yeah.

Lex Fridman 18:06 Like where you could pay for it easily. 'Cause like imagine...... I can imagine a lot of products that would just live as extensions, like improvements for social media, like-

Pieter Levels 18:15 Yeah.

Lex Fridman 18:16 ... a thing that was-

Pieter Levels 18:16 It's like GPTs, you know?

Lex Fridman 18:17 GPTs, yeah.

Pieter Levels 18:17 Like these ChatGPTs, they, they gonna charge money for it now, and you get a revu- rev share, I think, from OpenAI. I made a lot of them also.

Lex Fridman 18:24 Why? We'll, we'll talk about it. Let's, uh-

Pieter Levels 18:25 Yeah.

Lex Fridman 18:25 ... let, let's rewind back.

Pieter Levels 18:26 (laughs)

Lex Fridman 18:26 It's a pretty cool idea to do 12 startups in 12 months. What's, what's it take to build a thing in, in, in 30 days? Like at that time, how hard was that?

Pieter Levels 18:36 Mm. (sighs) I think it, the hard part is, like, figuring out what you shouldn't add, right, what you shouldn't build, because you don't have time. So you need to build a landing page. Y- Well, you need to make the pr-

Lex Fridman 18:45 (laughs)

Pieter Levels 18:46 No, you need to build the product, actually, 'cause there needs to be something to pay for it. Um, do you need to build a login system? Like, maybe no, you know? Like, maybe you can build some scrappy login system. Like, for Photoi, you sign up, you pay with Stripe checkout, and you get a login link. And when I started, there was only a login link with a hash, and that's just a static link, so it's very easy to log in.

Lex Fridman 19:03 Yeah.

Pieter Levels 19:04 It's not so safe, you know. What if you leak the link? And now I have real Google login, but that took, like, a year. So keeping it very scrappy is very important to... 'Cause you don't have time, you know? You need to focus on, um, what you can build fast.

Lex Fridman 19:16 Mm-hmm.

Pieter Levels 19:17 Uh, so money, Stripe, uh, build a product, build a landing page. Um, you need to think about how are people gonna find this. So are you gonna put it on Reddit or something?

Lex Fridman 19:26 Mm-hmm.

Pieter Levels 19:26 How are you gonna put it on Reddit without being looked at as a spammer, right? Like, um, if you say, "Hey, this is my new startup, you should use it," no. Nobody's... It, it gets deleted, you know?

Lex Fridman 19:34 Mm-hmm.

Pieter Levels 19:35 Um, maybe if you find a problem that a lot of people on Reddit already have on a subreddit, you know, like, and you solve that problem. Say, "Some people, I made this thing that might solve your problem, and maybe it's free for now," you know? Like, uh, that could work, you know. But you need to be very... You know, um, narrow it down-

Lex Fridman 19:52 Mm-hmm.

Pieter Levels 19:52 ... what you're building.

Lex Fridman 19:53 'Cause time is limited.

Pieter Levels 19:54 Yeah.

Lex Fridman 19:55 Actually, can we go back to the you laying in a room, feeling like a loser?

Pieter Levels 20:00 (laughs) Yeah.

Lex Fridman 20:00 I still feel like a loser sometimes. What's... What can you... Can you speak to that feeling, to that place, of just like feeling like a loser? And 'cause I think a lot of people in this world are laying in a room right now listening to this-

Pieter Levels 20:16 Yeah. Yeah, yeah.

Lex Fridman 20:17 ... and feeling like a loser.

Pieter Levels 20:18 Okay, so I think it's normal if you're young that you feel like a loser, first of all.

Lex Fridman 20:21 Especially when you're 27.

Pieter Levels 20:23 Yes. (laughs) Yeah, especially-

Lex Fridman 20:24 There's like a peak.

Pieter Levels 20:25 Yeah. Yeah, I think 27's the peak. And so I would not kill yourselves. It's very important. Just get through it, you know? But, uh, because you have nothing. You have n- probably no money. You have no business. You have no job. You ha- You... Y- Like, Jordan Peterson said this. I saw it somewhere. Like, the reason people are depressed is because they have nothing. They don't have a girlfriend. They don't have a, a boyfriend. They don't have... You need stuff. You need like a r- a family. You need things around you. You need to build a life for yourself. If you don't build a life for yourself, you'll be depressed. So if you're alone in Asia in a hostel, looking at the ceiling, and you don't have any m- money coming in, you don't have a girlfriend, you don't... Of course you're depressed. It's logic. But back then, if you're in the moment, you think, "This is not logic. Uh, there's something wrong with me," you know?

Lex Fridman 21:02 Yeah.

Pieter Levels 21:03 Um, and, and also, I think I started going... I started getting, like, anxiety, and I think I started going a little bit crazy, where I think travel can make you insane. And I know this because I know that there's like digital nomads that they kill themselves. And I don't, I haven't checked like this, the comparison with like baseline people, like suicide rate, but I have a hunch, um, especially in the beginning when it was a very new thing, like 10 years ago, that it can be very psychologically taxing. And you're alone a lot. S- Back then, when you travel alone, there was no other digital nomads back then, a lot. So you're in a strange culture. You look different than everybody. Like, you're in... I was in Asia. Like, you l- Everybody's really nice in Thailand, but you're not part of the culture. You're traveling around. You're hopping from city to city. You don't have a home anymore.

Lex Fridman 21:49 Mm-hmm.

Pieter Levels 21:50 You feel this routed.

Lex Fridman 21:51 And you're constantly an outcast in this, in that you're different from everybody else.

Pieter Levels 21:55 Yes, exactly. But people treat you... Like Thailand-

Lex Fridman 21:57 Yeah.

Pieter Levels 21:57 ... people are so nice, but you still feel like outcasts. And, and then I think that the digital nomads I met then were all kind of like... It was like shady business, you know? They were like vigilantes 'cause it was a new thing.

Lex Fridman 22:06 Mm-hmm.

Pieter Levels 22:07 And like one guy was selling illegal drugs. It was an American guy who was selling illegal drugs via UPS to Americans, you know, on his website. There were like a lot of dropshippers doing shady stuff. Um, there was a lot of shady things going on there. And they were, they didn't look like very balanced people. They didn't look like people I wanted to hang with, you know? So I also felt outcast from other foreigners in Thailand, other digital nomads. And I was like, "Man, I made a big mistake." And then I went back to Holland, and then I got even more depressed.

Lex Fridman 22:32 You said digital nomad. What is digital nomad? What is that way of life? What is the philosophy there? And the history-

Pieter Levels 22:38 So-

Lex Fridman 22:38 ... of the movement.

Pieter Levels 22:39 I, I, I struck upon it on accident, 'cause I, I was like, "I'm gonna graduate university, and then I'm gonna... I need to get out of here. I, I'll fly to Asia," 'cause I've been before in Asia. I studied in Korea in 2009 like study exchange. So I was like, "Asia easy, Thailand's easy. I'll just go there, figure things out." And it's cheap, it's very cheap. Chiang Mai, I would live like for $150 per month rent-

Lex Fridman 22:55 Yeah.

Pieter Levels 22:56 ... for like a private room, pretty good. So I struck up on this on accident. I was like, "Okay, there's other people on laptops working on their startup or working remotely." Back then, nobody worked remotely, but they worked on their, their businesses, right? Um, and they would, you know, live in like Columbia or, uh, Thailand or Vietnam or Bali. They would live kind of like in more cheap places, and it looked like a very adventurous life. Like, you travel around. You build your business. There's no pressure from like your home society, right? Like, you're American, so you're- you get pressure from American society telling you kind of what to do. Like, "You need to buy a house," or, "You need to do this stuff." I had this in Holland too. And you can get away from this pressure, and you can finally kind of feel like you're free. You're kinda... There's nobody telling you what to do. But that's also why you start feeling like you go crazy, 'cause you are, you are free. You're disattached from anything-

Lex Fridman 23:43 Mm-hmm.

Pieter Levels 23:43 ... and anybody. Um, you're disattached from your culture. You're disattached from the culture you're probably in 'cause you're staying very short.

Lex Fridman 23:49 I think Franz Kafka said, "I'm free, therefore I'm lost."

Pieter Levels 23:53 Man, that's so true. Yeah, that's exactly the point. And yeah, freedom is like... It's the definition of no constraints, right? Like, ev- anything's possible. You can go anywhere. And everybody's like, "Oh, that must be super nice," you know? Like, "Freedom, you must be very happy." And it's the opposite. Like, I don't think that makes you happy. I think constraints probably make you happy. And that's a big lesson I learned then.

Lex Fridman 24:14 But what, what were they making for money? What, what, so you're saying they were doing shady stuff at that time?

Pieter Levels 24:19 For me, you know, because I was more like a developer, I wanted to make startups kind of. And, and it was like, um, it was like drugs being shipped to America, like diet pills and stuff. Like, non-FDA approved stuff, you know? And they would like la- ... There was no like eth- ... They were like, they would, they would sit with beers, and they would laugh about like all the dodgy shit kinda they were doing, you know?

Lex Fridman 24:37 Ah, that part of the ... Okay, I see.

Pieter Levels 24:38 That kind of vibe, you know, like kind of sleazy e-com vibe. I'm not saying all e-com is sleaze, you know, but ... (laughs)

Lex Fridman 24:43 Right.

Pieter Levels 24:43 But, you know, this vibe.

Lex Fridman 24:44 It could be a vibe. And your, your vibe was more-

Pieter Levels 24:47 Mak- make cool stuff. Yeah.

Lex Fridman 24:47 ... build cool shit that's ethical.

Pieter Levels 24:50 You know the guys with sports cars in Dubai, these people, you know?

Lex Fridman 24:52 Yes.

Pieter Levels 24:53 E-com, like, "Oh, bro, you got to drop-ship."

Lex Fridman 24:55 Yeah.

Pieter Levels 24:55 "And you'll make 100 million a, a month." Like, those people, it was this shit. And I, I was like, "This is not my people."

Lex Fridman 25:01 Yeah. I don't ... I mean, there's nothing wrong with any of those individual components-

Pieter Levels 25:04 No, no judgment.

Lex Fridman 25:05 ... but, but there's a foundation that's not quite ethical. I think, what is that? I don't know what that is. But yeah, I get you.

Pieter Levels 25:12 No, I, like, I don't want to judge. It was more, I know that for me, it wasn't my world.

Lex Fridman 25:15 Yeah.

Pieter Levels 25:15 It wasn't my subculture.

Lex Fridman 25:17 Yeah.

Pieter Levels 25:17 I wanted to make cool shit, you know, but they also think their cool shit is cool, so, you know. Uh, but I wanted to make like real like startups, and that was my thing. I would read Hacker News, you know? Like-

Lex Fridman 25:26 Mm-hmm.

Pieter Levels 25:26 ... Y Combinator. And they were making cool stuff, so I wanted to make cool stuff.

Lex Fridman 25:30 I mean, that's a pretty cool way of life, just, if you romanticize it for a moment.

Pieter Levels 25:34 It's very romantic, man. It's very ... It's colorful, you know? Like, if I think about the memories.

Lex Fridman 25:38 I mean, what are some happy memories? Just like working, work, working in cafes or working in ... Just the freedom that, that, uh, envelops you from, with that, that way of life, 'cause anything is possible. You can just get up and go.

Pieter Levels 25:53 No, I think it was amazing. Like, we would work. Like, you would, I would make friends, and we would work until, you know, 6:00 AM in Bali, for example, with like, um, with Andre, my best friend, who I s- who's still my best friend. And we would, and other friends, and we would work until like the morning when the sun came up, because at night, the co-working space was silent, you know? There was nobody else. And I would wake up like 6:00 PM or 5:00 PM. I would drive to the co-working space on a motorbike. I would buy like, uh, 30 hot lattes from a cafe.

Lex Fridman 26:24 How many?

Pieter Levels 26:25 30, 'cause there was like s-

Lex Fridman 26:26 Wow.

Pieter Levels 26:26 ... there was like six people coming. We didn't know. Sometimes p- people would come in. And w-

Lex Fridman 26:30 Did you say three zero, 30?

Pieter Levels 26:32 Yeah.

Lex Fridman 26:33 Nice.

Pieter Levels 26:33 And we would drink like four per person or something, you know?

Lex Fridman 26:36 Yeah. Uh-huh.

Pieter Levels 26:36 Man, it's Bali. I don't know if they were powerful lattes, you know, but they were lattes. And, um, we'd put them in a plastic bag, and then we'd drive there, and all the coffee was like falling, you know, everywh-

Lex Fridman 26:45 Yeah.

Pieter Levels 26:45 And then we'd go in the co-working space and have these coffees here, and we'd work all night. We'd play like techno music, and everybody would just work in their ... Like, this was literally like business people. They would work on their startup, and we'd all try and make something. And then y- the sun would come up, and the morning people, the, you know, the, the yoga, uh, yoga girls and yoga guys would come in, you know, after the yoga class-

Lex Fridman 27:06 Yeah.

Pieter Levels 27:06 ... at 6:00. And they'd say, "Hey, good morning." And we're like, we look like this, you know? And we're like-

Lex Fridman 27:10 Yeah.

Pieter Levels 27:10 ... "What's up? How are you doing?" And we didn't know how bad we looked, you know, but it was very bad. And uh, then we would go home, sleep in like a hostel or a hotel, and do the same thing, and again and again and again. And it was this lock in mode, you know, like working. Um, and that was very fun.

Lex Fridman 27:26 So it was just a, a bunch of you, techno music blasting all through the ni- ... Yeah.

Pieter Levels 27:31 More like ... Like industrial.

Lex Fridman 27:33 Oh, so rapid bass.

Pieter Levels 27:33 Not like this cheesy-

Lex Fridman 27:35 I g- see, I got, for me, it, it's such an interesting thing, because the speed of the beat affects how I feel about a thing.

Pieter Levels 27:41 Yeah.

Lex Fridman 27:41 So the faster it is, the more anxiety I feel, but that anxiety's channeled into productivity. But if it's a little too fast, I start, the anxiety overpowers the-

Pieter Levels 27:50 So you don't like drum and bass music?

Lex Fridman 27:52 Probably not.

Pieter Levels 27:53 No, it's too fast.

Lex Fridman 27:54 The, I mean, for working. I have to, I have to-

Pieter Levels 27:56 Yeah.

Lex Fridman 27:56 ... play with it. It's like you can actually like, I can adjust my-

Pieter Levels 27:59 Yeah.

Lex Fridman 27:59 ... level of anxiety. There's a, must be a better word than anxiety. It's like a productive anxiety that I like.

Pieter Levels 28:07 Yeah.

Lex Fridman 28:07 Whatever that is.

Pieter Levels 28:07 And it also depends what kind of work you do, right? Like if you're writing, you probably don't want drum and bass music. I think for codes, like industrial techno, this kind of stuff, f- kind of fast, it works well, 'cause you really get like locked in. And combined with caffeine, you know, you, you go-

Lex Fridman 28:22 (laughs)

Pieter Levels 28:22 ... you go deep, you know? And-

Lex Fridman 28:24 Yeah.

Pieter Levels 28:24 ... and I think you balance on this edge of anxiety, because this caffeine is also hitting your anxiety. You want to be on the edge of anxiety with this techno running. Sometimes it gets too much. Like, stop the techno, stop the music, just like ... But, uh, but those are good memories, you know? And also like travel memories, like you go from city to city.

Lex Fridman 28:39 Yeah.

Pieter Levels 28:39 And it feels like, it's kind of like jet set life. Like, it's, it's feels very beautiful, like you're, you're seeing a lot of cool cities. And-

Lex Fridman 28:46 What was your favorite place that you remember you visited?

Pieter Levels 28:50 I think still like, uh, Bangkok is the best place. To e- and Bangkok and Chiang Mai. I think Thailand is very special. Like, I've been to the other places. Like, I've been to Vietnam, and I've been to South America and stuff. I still think Thailand wins in how nice people are, how easy of a life people have there.

Lex Fridman 29:08 Everything's cheap?

Pieter Levels 29:10 Yeah.

Lex Fridman 29:10 And good.

Pieter Levels 29:10 Well, Bangkok is getting expensive now. But Chiang Mai is still cheap. I think when you're starting out, it's a great place. Man, the air quality sucks. That's a big problem. So, um, and it's quite hot. But that's a very cool place. Um-

Lex Fridman 29:22 Pros and cons.

Pieter Levels 29:23 I love Brazil also.

Lex Fridman 29:25 Mm-hmm.

Pieter Levels 29:25 My girlfriend's Brazilian, but I do lo- ... I, not just because of that, but I like Brazil. Uh, the problem still is the safety issue, you know? Like, it, it's, it's like in America, like it's localized. It's hard for Europeans to understand like safety is localized to specific areas. So if you go to the right areas, it's amazing. Brazil is amazing. If you go to the wrong areas, like you, maybe you die.

Lex Fridman 29:42 Mm-hmm.

Pieter Levels 29:43 Right?

Lex Fridman 29:43 Yeah. Yeah, I mean, that's true.

Pieter Levels 29:45 But it's not true in Europe. So in Europe, it's much more eco-

Lex Fridman 29:47 That's true. That's, you're right.

Pieter Levels 29:48 More average.

Lex Fridman 29:49 You're right. You're right. There's, (laughs) it's more averaged out.

Pieter Levels 29:52 Yeah.

Lex Fridman 29:53 I like it when there's strong neighborhoods, when you're like, you cross a certain street, and you're in a dangerous part of town.

Pieter Levels 30:00 Man, yeah.

Lex Fridman 30:00 It's, I, I, I like it. I like, there's certain cities in the United States like that.

Pieter Levels 30:04 Yeah.

Lex Fridman 30:05 I like that. And you're saying Europe is more mellow.

Pieter Levels 30:06 What, you don't feel scared?

Lex Fridman 30:08 Well, I don't-

Pieter Levels 30:08 But you do-

Lex Fridman 30:09 I like danger.

Pieter Levels 30:09 ... BJJ.

Lex Fridman 30:11 No, not even just that. I think danger's interesting. So-

Pieter Levels 30:13 Yeah.

Lex Fridman 30:14 Danger reveals something about yourself, about others. Also, I like the full range of humanity.

Pieter Levels 30:20 Yeah.

Lex Fridman 30:20 So I don't like the mellowed out.... aspects of humanity.

Pieter Levels 30:23 I- I have friends, like these nomad friends that are exactly like this, like they go to, like, the- the kind of broken areas, you know? Like, they like this reality, they like the authenticity more. They don't like luxury, they don't like-

Lex Fridman 30:34 Oh, yeah, I hate luxury.

Pieter Levels 30:35 No. Yeah.

Lex Fridman 30:35 Yeah.

Pieter Levels 30:35 It's very European of you. Like... (laughs)

Lex Fridman 30:38 Wait, was that a...

Pieter Levels 30:39 (laughs)

Lex Fridman 30:39 That's a whole nother conversation.

Pieter Levels 30:41 Yeah.

Lex Fridman 30:41 So you, uh... (laughs) You, uh, quoted Freya Stark, quote, "To awaken quite alone in a strange town is one of the most pleasant sensations in the world."

Pieter Levels 30:52 Yeah.

Lex Fridman 30:53 Uh, do you remember a time you've awoken in a strange town and felt like that? We're talking about small towns or big towns, or...

Pieter Levels 31:00 Man, anywhere. I think I wrote it in some blog post, and like, it was a common thing when you would wake up, and this was like... 'Cause I have this website, I started a website about this digital nomads, like called nomadlist.com, and it was a community, so it was like 30,000 other digital nomads, 'cause I was feeling lonely, so I built this website and I stopped feeling lonely. Like, I started make- organizing meetups and making friends. And, um, and it was very common that people would say they would wake up and they would forget where they are.

Lex Fridman 31:26 Yeah.

Pieter Levels 31:27 Like for the first ha- half minute, and they had to look outside like, "Wh- where am I? Which country?" Which sounds really like privileged, but it's more like funny, like you literally don't know where you are because you're so disrooted. Um, but there's something... Man, it's like Anthony Bourdain, you know? There's something pure about this kinda vagabond travel thing, you know? Like, it's behind me, I think. I don't... Like, now I travel with my girlfriend, right? It's very different, but it is romantic, like memories of this kinda like vagabond, individualistic solo life. But the thing is, it didn't make me happy, but it was very cool, but it didn't make me happy, right? It made me anxious.

Lex Fridman 32:03 There's something about it that made you anxious. I don't know, I still feel like that. It's a cool feeling. It's scary at first, but then you realize where you are, and you, and you... I don't know, it's like you awaken to the possibilities of this place when you feel like that.

Pieter Levels 32:17 That's it. That's it.

Lex Fridman 32:17 It's like great. And it's even when you're doing just basic travel.

Pieter Levels 32:21 Yeah.

Lex Fridman 32:21 Like going to San Francisco or something else.

Pieter Levels 32:22 Yeah, you have like the novelty effect, like you're in a new place, like here things are possible, you know, you- you- you're, you don't get bored yet, and, and that's why people get addicted to traveling, you know?

Lex Fridman 32:33 Back

Lex Fridman 32:34 to startups. You wrote a book on, uh, how to do this thing, and, uh, gave a great talk on it, how to do startups. The book's called Make: Bootstrapper's Handbook.

Pieter Levels 32:44 Yeah.

Lex Fridman 32:44 I was wondering if you could go through some of the steps. It's idea, build, launch, grow, monetize, automate, and exit. There's a lot of fascinating ideas in each one. So, idea stage.

Pieter Levels 32:55 Yeah.

Lex Fridman 32:55 How do you find a good idea?

Pieter Levels 32:56 So, I think you need to be able to spot problems. So for example, you can go in your daily life, like when you wake up, and you're like, "What are stuff that I'm really annoyed with, that's like, um, in my daily life that doesn't function well?" And that's a problem that you can see, "Okay, maybe that's something I can add, write code about, you know, code for, and it will make my life easier." So I would say make like a list of all these problems you have, and like an idea to solve it, and then see which one is like viable, you can actually do something, and then start building it.

Lex Fridman 33:25 So that's a, that's a really good place to start. Become open to all the problems in your life. Like, actually start noticing them.

Pieter Levels 33:33 Yeah.

Lex Fridman 33:33 I think that's actually not a trivial thing to do, to realize that some aspects of your life could be done way, way better.

Pieter Levels 33:39 Yeah.

Lex Fridman 33:39 'Cause we kinda very quickly get a- accustomed to discomforts.

Pieter Levels 33:44 Exactly.

Lex Fridman 33:45 Uh, like for example, like doorknobs.

Pieter Levels 33:47 Yeah.

Lex Fridman 33:48 Like design of certain things. (laughs) Like-

Pieter Levels 33:51 New Lex Freeman doorknob.

Lex Fridman 33:52 (laughs) Yeah.

Pieter Levels 33:52 $50.

Lex Fridman 33:52 Well, now, th- that one I know how much, uh, incredible design work has gone into. It's, uh, really interesting.

Pieter Levels 33:59 Yeah.

Lex Fridman 33:59 Doors and doorknobs, just the design of everyday things.

Pieter Levels 34:03 Yeah.

Lex Fridman 34:03 Forks and spoons. It's gonna be hard to come up with a fork that's better than the current-

Pieter Levels 34:07 Yeah.

Lex Fridman 34:07 ... fork designs.

Pieter Levels 34:09 Yeah.

Lex Fridman 34:09 And the other aspect of it is you're saying, like, in order to come up with interesting ideas, you gotta try to live a more interesting life.

Pieter Levels 34:15 Yeah. But that's where travel comes in, because when I started traveling, I started seeing stuff in other countries that you didn't have in Europe, for example, or America even. Like if you go to Asia, uh, like, dude, especially 10 years ago, nobody knew about this. Like WeChat, all these apps that they already had before we had them, these everything apps, right? Now Elon's trying to make X this everything app, like WeChat, same thing. Like in Indonesia or Thailand, you have one app that you can order food with, you can order groceries, you can order massage, uh, you can order car mechanic. Um, anything you can think of is in the app. And that stuff, for example, you, you know, that's called like arbitrage. You can go to, back to your country and build that same app for your country, for example. So you start seeing, um, problems, you start seeing solutions that other countries already, other people already did in the rest of the world. And also traveling in general just gives you more problems, 'cause travel is uncomfortable, you know? Um, airports are horrible. Airplanes are not comfortable either. Th- there's a lot of problems you start seeing just getting out of your house, you know?

Lex Fridman 35:19 But also you can... I mean, in the digital world, you can just go into different communities and see what can be improved by the-

Pieter Levels 35:25 Yes.

Lex Fridman 35:25 ... others in that.

Pieter Levels 35:26 Yeah, yeah, yeah. Yeah.

Lex Fridman 35:28 What specifically is your process of generating ideas? Do you, like, do idea dumps? Like do you have a document where you just keep writing stuff?

Pieter Levels 35:34 Yeah, I used to have like a... 'Cause when I was, when I wasn't making money, I was trying to, like, make this list of ideas to see, like, so I need to build... I was thinking statistically already, like I need to build all these things, and one of these will work out probably, you know? So I need to have a lot of things to try. Um, and I did that. Right now, I think, like, because I already have money, I can do more things, uh, based on technology. So for example, AI, when I found out about... when stable diffusion came, or, uh, ChatGPT and stuff, all this thing, all these things were like... I didn't start working with them because I had a problem. I had no problems, but I was very curious about technology, and I was, like, playing with it, and s- figuring out, like... First just playing with it, and then, and then you find something, like, okay, this generates, uh, stable diffusion generates houses very beautiful, and interiors, you know?

Lex Fridman 36:21 So it's less about problem solving and it's more about the possibilities of new things you can create

Narrator 36:25 Yeah.

Pieter Levels 36:25 ... with- But that's very risky, because that's the famous, like, solution trying to find a problem.

Lex Fridman 36:29 Yeah.

Pieter Levels 36:30 ... and usually it doesn't work. And that's very common with, with startup founders, I think. They, they have tech, but actually people don't need the tech, right?

Lex Fridman 36:37 Uh, can you actually explain... It'd be cool to talk about some of the stuff you've created. Can you explain, um, this, the photoai.com?

Pieter Levels 36:45 Yeah. Yeah. So it's like fire your photographer. The idea is like, you don't need a photographer anymore. You can train yourself as an AI model, and then you can take as many photos as you want anywhere in any clothes, uh, with ex- facial expressions like happy or sad, or, uh, poses, uh, all this stuff.

Lex Fridman 37:02 So how does, how does it work?

Pieter Levels 37:03 Yeah.

Lex Fridman 37:03 This is, uh, (laughs) you sent me-

Pieter Levels 37:06 So you could press-

Lex Fridman 37:06 ... a, a link to a gallery of, uh, ones done on me-

Pieter Levels 37:09 Yeah.

Lex Fridman 37:09 ... which is-

Pieter Levels 37:09 So on the left you have the prompts, the-

Lex Fridman 37:11 Yep.

Pieter Levels 37:11 ... box. Yeah. So you can write like... So model is your model. This is Lex Fridman.

Lex Fridman 37:14 Yeah. Yeah.

Pieter Levels 37:14 So you can write like, "Model as a blah blah blah," whatever you want.

Lex Fridman 37:18 Yep.

Pieter Levels 37:18 Then press the button and it will take photos. Take, we'll take like one minute.

Lex Fridman 37:21 60 photos. What are you using for the hosting for the compute?

Pieter Levels 37:24 Replicate.

Lex Fridman 37:25 Okay.

Pieter Levels 37:25 Replicate.com. They're very, very good.

Lex Fridman 37:27 Okay. It's cool like this, I- interface wise, it's cool that you're showing how long it's gonna take. This is amazing. So it's taken a, um... Presuming you just loaded in a few pictures from the internet.

Pieter Levels 37:37 Yeah. So I d- I went to Google Images, typed in Lex Fridman. I added like 10 or 20 images. You can open 'em in the gallery, and you can use your cursors to... Yeah. So some don't look like you. So the, the hit and miss rate is like-

Lex Fridman 37:50 Mm-hmm.

Pieter Levels 37:50 ... I don't know, I'd say like 50/50 or something. It's getting better.

Lex Fridman 37:53 But when I was watching your tweets, like it's been getting better and better and better.

Pieter Levels 37:56 It was very bad in the beginning. It was so bad, but still people signed up to it, you know? Uh...

Lex Fridman 38:01 (laughs) There's, there's two Lexes, it's great. It's getting more and more sexual, and it's making me-

Pieter Levels 38:07 Yeah, yeah, yeah.

Lex Fridman 38:07 ... very uncomfortable.

Pieter Levels 38:08 Man, but that's the problem with these models 'cause-

Lex Fridman 38:09 (laughs)

Pieter Levels 38:10 No, we need to talk about this 'cause the models-

Lex Fridman 38:12 Sure. (laughs)

Pieter Levels 38:12 ... with diffusion-

Lex Fridman 38:13 Yeah.

Pieter Levels 38:13 ... so the photorealistic models that are like fine-tuned-

Lex Fridman 38:16 Yeah.

Pieter Levels 38:16 ... they were all trained on porn in the beginning, and there was a guy called Hassan. So I was trying to figure out how to do photorealistic AI photos, and it was... Stable diffusion by itself is not doing that well. Like the faces look all mangled.

Lex Fridman 38:27 Yep.

Pieter Levels 38:28 Um, and it doesn't have enough resolution or something to, to, to do that well. So... But I started seeing these, these, m- um, base models, these fine-tuned models and the people who train on porn, and I would try them, and they would be very photorealistic. They would have bodies that actually made sense, like body au- uh, a- anatomy. Um, but if you look at the photorealistic models that people use now still, there's still a core of porn there, like of naked people.

Lex Fridman 38:51 Yeah.

Pieter Levels 38:51 So I need to prompt out the nak- and everyone needs to do this with AI startups with imaging. You need to prompt out the naked stuff. You need to put a, mm, you know, nakeds, um-

Lex Fridman 38:59 You have to keep reminding the model, "You need to put-"

Pieter Levels 39:02 It's not-

Lex Fridman 39:02 ... clothes on the thing."

Pieter Levels 39:02 Yeah. Don't put nakeds 'cause it's very risky. I have Google Vision that checks-

Lex Fridman 39:05 Yeah.

Pieter Levels 39:05 ... every photo before it's shown to the user to like check for NSFW.

Lex Fridman 39:08 Like a nipple detector?

Pieter Levels 39:09 Yes.

Lex Fridman 39:09 Oh, NSFW (laughs) d- detector.

Pieter Levels 39:11 Because you get the journalists get very angry if they, you know?

Lex Fridman 39:14 If you sexualize.

Pieter Levels 39:15 There was a journalist, I think, that w- got angry that used this and was like, "Oh, it, it made me... It showed like a nipple." Because Google Vision didn't detect it. So, that's like... These are the kind of problems you need to deal with, you know? That's what I'm talking about.

Lex Fridman 39:26 (laughs)

Pieter Levels 39:26 This is with cats. But look at the cat face. It's also kind of mangled.

Lex Fridman 39:31 (laughs)

Pieter Levels 39:31 You know?

Lex Fridman 39:32 I'm, I'm, I'm, uh, I, I'm a little bit disturbed.

Pieter Levels 39:36 You can zoom in on the cat if you want, like, like...

Lex Fridman 39:39 Yeah. (laughs)

Pieter Levels 39:39 Yeah. It's a very sad cat. It doesn't have a nose.

Lex Fridman 39:42 It doesn't have a nose.

Pieter Levels 39:44 But this... Wow. Man, but this is the problem with AI startups, 'cause they all act like it's perfect, like this is groundbreaking and...

Lex Fridman 39:49 Yeah.

Pieter Levels 39:49 But it's not perfect. It's like really bad, you know, half the time.

Lex Fridman 39:53 So if I wanted to sort of update model as...

Pieter Levels 39:55 Yeah. So you remove this stuff and you write like whatever you want, like in Thailand or something, or in Tokyo.

Lex Fridman 40:02 Uh, in Tokyo?

Pieter Levels 40:04 Yeah.

Lex Fridman 40:06 And then...

Pieter Levels 40:07 You could say like, "At night with neon lights." Like, you can add more detail to make it more-

Lex Fridman 40:11 So I'll, I'll go in Austin. Do you think it'll know?

Pieter Levels 40:13 Yeah. Austin.

Lex Fridman 40:13 In, in Texas? In Austin, Texas?

Pieter Levels 40:14 With cowboy hat.

Lex Fridman 40:16 In Texas. Yeah.

Pieter Levels 40:17 As a, as a cowboy.

Lex Fridman 40:21 As a cowboy. It's gonna go so, uh, towards the porn direction. It's, it's, it's gonna-

Pieter Levels 40:25 Man, I hope not. This is the end of my career. (laughs)

Lex Fridman 40:27 (laughs) Or the beginning. It depends. We can send you a push notification when your photos are done.

Pieter Levels 40:32 Yeah.

Narrator 40:32 Yeah. Okay.

Lex Fridman 40:32 All right, cool.

Pieter Levels 40:34 Yeah, let's see.

Lex Fridman 40:35 Oh, wow. So this whole interface you've built?

Pieter Levels 40:37 Yeah.

Lex Fridman 40:38 This is really well done.

Pieter Levels 40:39 It's all jQuery.

Lex Fridman 40:40 (laughs)

Pieter Levels 40:40 Do I still use jQuery?

Lex Fridman 40:42 Yes.

Pieter Levels 40:42 I'm the only one?

Lex Fridman 40:42 Still.

Pieter Levels 40:43 After 10 years?

Lex Fridman 40:43 To this day. You're not the only one. The entire world, the web-

Pieter Levels 40:46 Yeah.

Lex Fridman 40:46 ... is PHP. The stack is-

Pieter Levels 40:47 It's PHP and jQuery. Yeah. And SQLite.

Lex Fridman 40:50 You're just like one of the top performers from a programming perspective that are still like openly talking about it. But ev- everyone's using PHP. Like if you look at mo- most of the web, it's still probably PHP and jQuery.

Pieter Levels 41:02 It's, I think, 70%. It's 'cause of WordPress, right? 'Cause the blogs are-

Lex Fridman 41:04 Yeah, that's true.

Pieter Levels 41:05 Yeah. Um...

Lex Fridman 41:06 That's true.

Pieter Levels 41:06 I'm seeing a revival now. People are getting sick of frameworks. Um, like all the JavaScript frameworks are so like, what do you call it? Like wieldy. Like they're so... It takes so much work to just maintain this code. And then it updates to a new version. You need to change everything. PHP just stays the same and works.

Lex Fridman 41:23 Yeah.

Pieter Levels 41:24 And...

Lex Fridman 41:24 Can you actually d- just speak to that stack? You u- so you build all your websites, apps, startups, projects, all of that with mostly vanilla HTML?

Pieter Levels 41:33 Yeah.

Lex Fridman 41:34 Uh-

Pieter Levels 41:35 CSS.

Lex Fridman 41:35 ... JavaScript with J- jQuery, PHP, and, uh, SQLite.

Pieter Levels 41:39 (laughs)

Lex Fridman 41:39 Um, and so that, that's a really simple stack.

Pieter Levels 41:43 Yeah.

Lex Fridman 41:43 And you get stuff done really fast with that. Can you just speak to the philosophy behind that?

Pieter Levels 41:47 I think it's accidental 'cause that's the thing I, I knew. Like, I knew PHP, I knew HTML, um, CSS, you know, because you make websites. And when my startup started taking off, I didn't have time to... I remember putting on my to-do list like, "Learn Node.js," 'cause I, it's important to switch, you know? 'Cause-

Lex Fridman 42:04 Yeah.

Pieter Levels 42:04 ... this obviously is much better language than PHP. And I never learned it. I never did it, uh, 'cause I didn't have time. I, I, these things were growing like this, and I was launching more projects and I never had time. It's like one day, you know, I'll, I'll start coding properly, and I never got to it.

Lex Fridman 42:19 I, I sometimes wonder if I need to learn that stuff. It's still a to-do item for me to really learn Node.js or, or Flask, or these kind of the-

Pieter Levels 42:27 React, Vue.

Lex Fridman 42:28 Yeah. React.

Pieter Levels 42:30 Yeah.

Lex Fridman 42:30 And it's just... There's-It feels like a responsible software engineer should know how to use these.

Pieter Levels 42:38 Yeah.

Lex Fridman 42:38 But you can get stuff done so fast-

Pieter Levels 42:42 It's different, right?

Lex Fridman 42:42 ... with vanilla versions of stuff.

Pieter Levels 42:44 Yeah. It's like software developers, if you want to get a job, and there's like, you know, people making stuff, like startups. And if you want to be an entrepreneur, probably you sh- maybe shouldn't, right?

Lex Fridman 42:53 I wonder if there's like... I really wanna measure performance and speed. I think there's a deep wisdom in that.

Pieter Levels 43:00 Yeah.

Lex Fridman 43:00 I do think that frameworks and just constantly wanting to learn the new thing, this complicated way of software engineering gets in the way. I'm not sure what to say about that, because definitely like you shouldn't build everything from just vanilla JavaScript or vanilla C, for example.

Pieter Levels 43:16 Yeah.

Lex Fridman 43:16 C++ when you're building systems engineering is like, there's a lot of benefits for p- uh, pointer safety and all that kind of stuff. So, I, I don't know, but it just feels like you can get so much more stuff done if you don't care about how you do it.

Pieter Levels 43:33 Man, this is my most controversial take, I think, and maybe I'm wrong, but I feel like... There's frameworks now that raise money, they raise a lot of money, like they raise 50 million, 100 million, $300 million. And the idea is that you need to make the developers, the new developers, like when you're 18 or 20 years old, right? Um, get them to use this framework and then add a platform to it, like where the framework can... It is open source, but you probably should use the platform, which is paid, to use it. And the, the cost of the platforms to host it-

Lex Fridman 44:05 Mm-hmm.

Pieter Levels 44:06 ... are 1,000 times higher than just hosting it on a simple AWS server or a VPS on DigitalOcean, right? Um, so there's obviously like a monetary incentive here. Like we wanna get a lot of fe- developers to use this technology, and then we need to charge them money 'cause they're gonna use it in startups, and then the startups can pay for the bills.

Lex Fridman 44:22 Mm-hmm.

Pieter Levels 44:23 Um, but what that k- it kinda destroys the, the information out there about learning to code, because they, you know, they pay YouTubers, they pay influencers. Developer influencer is a big thing to like, um... And the same thing what happens with like, uh, nutrition and fitness or something, same thing happens in developing. They pay this influencer to promote the stuff, use it, make stuff with it, make demo products with it. And then a lot of people are like, "Wow, uh, use this." And I started noticing this, 'cause when I would ship my stuff, people would ask me, "What are you using?" I would, I would say, "Just PHP, jQuery, why does it matter?" And people would start kinda attacking me, like, "Why are you not using this new technology, this new framework, this new thing?" And I say, "I don't know, 'cause this PHP thing works, and I don't really, I'm optimizing for anything, I'm just doing, it just works." And I never understood like why, uh, like I understand there's new technologies that are better, uh, and there should be improvement. But I'm very suspicious of money, just like lobbying. There's money in this developer framework scene. There's hundreds of millions-

Lex Fridman 45:22 Mm-hmm.

Pieter Levels 45:22 ... that goes to ads or influencers, whatever. It can't all go to developers. You don't need so many developers for a framework. And it's open source. Do make a lot of more money on these startups.

Lex Fridman 45:32 So that's a really good perspective. But in, in addition to that is like when you say better, it's like, can we get some data on the better?

Pieter Levels 45:42 (laughs)

Lex Fridman 45:42 Because like, I wanna know from the individual developer perspective, and then from a team of five, team of 10, team of 20 developers.

Pieter Levels 45:49 Yeah.

Lex Fridman 45:51 Uh, measure how productive they are in, uh, shipping features, how many bugs they create, how many security holes, uh, results-

Pieter Levels 46:00 PHP was not good at the security for a while, but now it's good, yeah.

Lex Fridman 46:02 In theory.

Pieter Levels 46:03 Yeah.

Lex Fridman 46:03 In theory. Is it though? I'd l-

Pieter Levels 46:05 Now it's good.

Lex Fridman 46:06 No. Uh, no, no, now as, as you're saying it, I wanna know if that's true, because PHP was just the majority-

Pieter Levels 46:14 Ah, yeah.

Lex Fridman 46:14 ... of websites on the internet.

Pieter Levels 46:15 Could be true.

Lex Fridman 46:16 Uh, uh, is, is it just overrepresented? Same with WordPress.

Pieter Levels 46:19 Yeah.

Lex Fridman 46:19 Yes, there's a reputation that WordPress has a gigantic number of security holes. I don't know if that's true. I know, I know it gets attacked a lot, because it's so popular.

Pieter Levels 46:28 Yeah.

Lex Fridman 46:28 It definitely does have security holes, but s- maybe a lot of other systems have security holes as well. Anyway, I, I just was sort of questioning the conventional wisdom that keeps wanting to push software engineers towards frameworks, towards complex-

Pieter Levels 46:41 Completely, yeah. Yeah, yeah, yeah.

Lex Fridman 46:42 ... like, uh, super complicated sort of software engineering approaches that stretch out the, the time it takes to actually build the thing.

Pieter Levels 46:50 Man, 100%. And it's the same thing with big corporations, like 80% of the people don't do anything. It's like-

Lex Fridman 46:55 Right.

Pieter Levels 46:55 ... it's not efficient. And, and if you, if your, your, your benchmark is like, uh, people building stuff that actually gets done and like for society, right? Like if we wanna save time, we should probably use technologies, um, th- that's simple, that's pragmatic, that's like, uh, that works, uh, that's not overly complicated, doesn't make your li- life like a living hell, you know? And-

Lex Fridman 47:18 And use a framework when it obviously solves a problem, a direct problem-

Pieter Levels 47:22 Yeah.

Lex Fridman 47:22 ... that you-

Pieter Levels 47:22 Of course, you, yeah, of course. I'm not saying you should code without a framework. I'm, you should use whatever you want, but, yeah, I think it's suspicious, you know? And, um-

Lex Fridman 47:31 (laughs)

Pieter Levels 47:31 And I think it's suspicious, when I talk about it on Twitter, pe- like there's a lot of, there's, this army comes out, you know?

Lex Fridman 47:35 Yeah.

Pieter Levels 47:35 There's this, these framework armies.

Lex Fridman 47:37 Yeah.

Pieter Levels 47:37 It's, man, something my gut tells me.

Lex Fridman 47:40 I wanna ask the framework army what have they built this week. It's the Elon question. What'd you do this week?

Pieter Levels 47:45 Yeah, and did you make money with it, you know? Did you charge users? Is it a real business? And, um, yeah.

Lex Fridman 47:51 Uh, so going back to the cowboy. First of all, every, every-

Pieter Levels 47:55 Some don't look like you, right? But some do.

Lex Fridman 47:56 Every aspect of this is pretty incredible. I'm also just looking at the interface. This is really well done. So this is all just jQuery and-

Pieter Levels 48:02 Yeah.

Lex Fridman 48:02 This is really well done. So like, tell m- t- take me through the journey of PhotoAI. Like, you don't know mu- most of the world doesn't know much about, uh, Stable Diffusion or any of this-

Pieter Levels 48:13 Yeah.

Lex Fridman 48:14 ... a- any of this generative AI stuff. And s- so you're thinking, "Okay, how can I build cool stuff with this?"

Pieter Levels 48:19 Yeah.

Lex Fridman 48:19 So, what was the origin story of PhotoAI?

Pieter Levels 48:21 I think, uh, it, it started 'cause Stable Diffusion came out. So Stable Diffusion's like-

Lex Fridman 48:25 Yeah.

Pieter Levels 48:25 ... this, the first like generative image model, AI model. And, um, I started playing with it, like you could install it on your Mac. Like somebody forked it and made it work for MacBooks. So I downloaded it and, uh, cloned the repo and started using it to generate images. Um, and-... it was, like, amazing. Like, it would, I found it on Twitter, because you see things happen on Twitter, and I, I would post what I was making on Twitter as well, and you could make any image. You could write a prompt. So essentially, you write a prompt, and then it generates a photo of that, or an image of that, um, in any style. Like, they would use, like, artist names to make, like, a Picasso kind of style and stuff. Um, and I was trying to see, like, what is it good at? Is it good at people? No, it's really bad at people. But it was good at houses. So architecture, for example. I would generate, like, architecture houses. Um, so I made a website called thishousedoesnotexist.org, and it generated, like... They called it, like, house porn in that one.

Lex Fridman 49:22 (laughs)

Pieter Levels 49:23 Like, house porn, it's like a subreddit. So... And this was Stable Diffusion, like, the first version, so it looked really... And you can click for another photo. So it generates, like, all these kind of non-existing houses.

Lex Fridman 49:34 It is house porn.

Pieter Levels 49:35 But it looked kind of good, you know? Like, especially back then.

Lex Fridman 49:37 It looks really good.

Pieter Levels 49:38 And now, now things look much better. Um...

Lex Fridman 49:42 That's really, really well done. Wow.

Pieter Levels 49:46 And it also generates, like, a description.

Lex Fridman 49:50 And you can upvote. Is it nice?

Pieter Levels 49:52 Yeah.

Lex Fridman 49:52 Upvote it.

Pieter Levels 49:52 Yeah.

Lex Fridman 49:52 Man, there's so much to talk to you about. Like, the choices here-

Pieter Levels 49:56 (laughs)

Lex Fridman 49:56 ... is really well done.

Pieter Levels 49:57 This was very scrappy. In the bottom, there's, like, a ranking of the most upvoted houses, so these are the top voted. And if you go to all time, you see quite beautiful ones. Yeah, so this one is my favorite, the number one. It's like, kind of like a, um...

Lex Fridman 50:10 How is this not more popular?

Pieter Levels 50:12 It was really popular for, like, a while, but then people got so bored of it, I think 'cause I was getting bored of it too. Like just-

Lex Fridman 50:19 Interesting.

Pieter Levels 50:19 ... continuous house porn. Like, everything starts looking the same.

Lex Fridman 50:22 Yeah.

Pieter Levels 50:22 But then, uh, I saw it was really good at interior, so I pivoted to interior-ai.com, where, um, I tried to, like, upload first generated interior designs, and then I tried to do, like there was a new technology called image to image, where you can input an image, like a photo, and it would kind of modify the thing.

Lex Fridman 50:41 Mm-hmm.

Pieter Levels 50:42 So you see, it looks almost the same as PhotoAI. It's the same code, essentially. Um...

Lex Fridman 50:46 Nice.

Pieter Levels 50:47 So I would, I would upload a photo of my interior where I lived, and I would ask, like, change this into, like, um, I don't know, like, maximalist design, you know? And it worked, and it worked really well. So I was like, "Okay, this is a startup," 'cause obviously interior design AI, and nobody was doing that yet. So I launched this, and it was successful and made, like, in, within a week, made 10K, 20K a month.

Lex Fridman 51:09 Mm-hmm.

Pieter Levels 51:10 And now it still makes, like, 40K, 50K a month, uh, and it's been like two years. So then I was like, "How can I improve this interior design? I need to start learning fine tuning." And fine tuning is where you have a existing AI model, and you fine tune it on a specific goal you want it to do. So I would find really beautiful interior design, make a gallery, and, um, train a new model that was very good at interior design. And it worked, and I used that as well. And then for fun, I uploaded photos of myself, (laughs) and here's where it happened, uh, and to train myself, like, and this would never work obviously, and it worked, and actually d- it started understanding me as a concept.

Lex Fridman 51:46 Mm-hmm.

Pieter Levels 51:46 So my face worked, and, and you could do, like, different styles, like me as a, like very cheesy, me-medieval warrior, all this stuff. So I was like, "This is another startup." So now I did avatar-ai.me. I couldn't get the .com. And this was, uh...

Lex Fridman 52:02 Is that still up?

Pieter Levels 52:02 This was a... Yeah, avatar-ai.me. Well, now it forwards to PhotoAI, 'cause it pivoted.

Lex Fridman 52:06 Got it.

Pieter Levels 52:07 But this was more, like, cheesy things. So this is very interesting, 'cause this went so viral. It made like, I think like, 150K in a week or something.

Lex Fridman 52:16 Mm-hmm.

Pieter Levels 52:16 It's the most money I ever made. And, um, and then big, this is very interesting, the big VC companies like Lensa, which are much better at iOS and stuff than me, I didn't have iOS app, they quickly built an iOS app that does the same, and they found technology. And it's all open technology, so it's good. And I think they made like $30 million with it.

Lex Fridman 52:35 Yeah.

Pieter Levels 52:35 They, they became, like, the top grossing app after that. Um, and it was-

Lex Fridman 52:40 What, how- how do you feel about that?

Pieter Levels 52:42 I think it's amazing, honestly. And it's not like-

Lex Fridman 52:44 You didn't have, like, a feeling like, "Oh, fuck. What was that?"

Pieter Levels 52:45 No. I was a little bit, like, sad, 'cause all my products would work out, and I never had, like, real fierce competition. And now I have, like, fierce competition from, like, a, a very skilled, high talent, like, iOS developer studio, someone that... And, and they already had a app. They had a app in app store for, like, um, I think retouching your face or something. So they were very smart. They add these avatars to there. It's a feature. They had the users just do a push notification to everybody, "We have these avatars."

Lex Fridman 53:10 Yeah.

Pieter Levels 53:11 Man, they made great, I think they made, they made so much money. And, um, I think they did a really great job. And, uh, I also made a lot of money with it. But that was... I- I- I quickly realized it wasn't my thing, 'cause it was so cheesy. It was like kitsch, you know?

Lex Fridman 53:24 Mm-hmm.

Pieter Levels 53:24 It was kind of like, um, me as a Barbie, or me as a, you know? It was-

Lex Fridman 53:28 Yeah.

Pieter Levels 53:29 ... too cheesy. I wanted to go for, like, what's a real problem we can solve, because this is gonna be a hype. This is gonna be... And it was a hype, these avatars. Um, so like, let's do real photography. Like, how can you make people look really photorealistic? And that was difficult, and that's why these avatars worked, 'cause they were all, like, in a cheesy, you know, Picasso style. And art is easy, 'cause you interpret the, the... The, all the problems that AI has with your face are, like, artistic, you know, if you call it Picasso. But if you make a real photo, all the problems with your face, like, it just, you look wrong, you know?

Lex Fridman 53:58 Mm-hmm.

Pieter Levels 53:59 So I started making PhotoAI, which was like a pivot of it, where it was like a photo studio, um, where you could take photos without actually needing a photographer, needing a studio. You don't, just, you know, you just type it. And I've been working on it for, like, the last year.

Lex Fridman 54:14 Yeah, it's really incredible. That journey is really incredible. Let's go to the beginning of PhotoAI though, 'cause I remember seeing a lot of really (laughs) really hilarious photos. I think you were using yourself as a case study, right?

Pieter Levels 54:27 Yeah.

Lex Fridman 54:27 Yeah, so what, uh... There's a tweet here. "Sold $100,000 in AI generated avatars."

Pieter Levels 54:36 Yeah, and it's a lot. Like, it's a lot for anybody. It's a lot for me. Like, uh-

Lex Fridman 54:39 (laughs)

Pieter Levels 54:40 ... making 10K a day on this, you know?

Lex Fridman 54:42 That's amazing.That's amazing.

Pieter Levels 54:46 And then the nested tweet, like that's the launch tweet. And then the- before there is like the- me hacking on it.

Lex Fridman 54:53 Oh, I see. (laughs) So that... Okay, so October 26th, 2022.

Pieter Levels 54:59 Yeah.

Lex Fridman 55:00 "I train an ML model on my face." (laughs) Sh-

Pieter Levels 55:05 Because my eyes are quite f- far apart, I learned when I did YouTube, I would put like a photo of like my DJ photo, you know, my makeup on. People would say I look like a hammerhead shark.

Lex Fridman 55:14 (laughs)

Pieter Levels 55:14 It was like the top comment. So then I realized my eyes are far apart. Uh...

Lex Fridman 55:17 Yeah. The internet helps you figure out what you look like.

Pieter Levels 55:19 Yeah, it helps you realize who y- how you look, you know?

Lex Fridman 55:21 Boy, do I love the internet.

Pieter Levels 55:23 That's my first, first trip.

Lex Fridman 55:26 Well, what is, is this... Wait.

Pieter Levels 55:27 It's water from the waterfall. But the waterfall's in the back, you know, so what's going on?

Lex Fridman 55:33 So this is, how much of this is real?

Pieter Levels 55:35 It's all AI.

Lex Fridman 55:36 It's all AI?

Pieter Levels 55:38 Yeah.

Lex Fridman 55:39 That's pretty good, though, for the early days.

Pieter Levels 55:40 Exactly. So, but this was hit or miss. So you had to do a lot of curation, 'cause 99% of it was really bad.

Lex Fridman 55:45 Mm-hmm.

Pieter Levels 55:46 So these are the photos I uploaded.

Lex Fridman 55:47 How many photos did you use? "Only these. I will try more up-to-date pics later." These are the- th- these are the only photos you uploaded?

Pieter Levels 55:55 Yeah. Yeah.

Lex Fridman 55:56 Wow. Wow. Okay, so like you were learning all this super quickly. What, uh-

Pieter Levels 56:03 Yeah.

Lex Fridman 56:03 W- what are some, like, interesting details you remember from that time, for like what you had to figure out to make it work? And for people just listening, he uploaded just a, just a handful of photos that don't really have a good capture of the face. And he's able to already-

Pieter Levels 56:16 I think it's cropped. It's like a crop out of, out of the layout. But they're, they're square photos, so they're 512 and 512.

Lex Fridman 56:21 Mm-hmm.

Pieter Levels 56:22 Because that's the only fusion. Um...

Lex Fridman 56:24 But nevertheless, not great capture of the face.

Pieter Levels 56:27 Yeah.

Lex Fridman 56:27 I mean, like, it's not, it's not like a, uh, collection of several hundred photos there is like...

Pieter Levels 56:34 Exactly, like I would, I would imagine that too when I started.

Lex Fridman 56:36 360 view.

Pieter Levels 56:36 I was like, "Oh, this must be like some 3D scan technology," right?

Lex Fridman 56:39 Yeah.

Pieter Levels 56:39 So I think the cool thing with AI, it trains the concept of you. So it's literally like learning, just like any AI model learns, it learns how you look.

Lex Fridman 56:47 Mm-hmm.

Pieter Levels 56:47 So I, I did this, and then I was getting so much... I mean, I was getting DMs, like Telegram messages, like, "How can I do the same thing? I want these photos."

Lex Fridman 56:55 Mm-hmm.

Pieter Levels 56:55 "My girlfriend wants these photos." So I was like, "Okay, this is obviously a business." But I didn't have time to code it, uh, make a whole, like, app about it. So I made an HTML page.

Lex Fridman 57:05 Mm-hmm.

Pieter Levels 57:06 Um, registered the domain name, and this was not even... It was a Stripe payment link, which means you have, uh, literally a link to Stripe to pay, but there's no code in the back. So all you know is you have customers that paid money. And then I added like, uh, some, uh, a Typeform link. So Typeform is a site where you can ha- make, create like your own, um, input form, like Google Forms. So they would get an email with a link to the Typeform, or actually just the link after the checkout, and they could upload their photos. So enter their email, upload the photos, and then, um, and I launched it, and I was like, "Here, first sale." This was October 2022. And I think within like the first 24 hours, it was like, I'm not sure, it was like 1,000 customers or something. Um, but the problem was I didn't have code to automate this, so I had to do it manually.

Lex Fridman 57:52 Mm-hmm.

Pieter Levels 57:52 So the first few hundred, I just literally took their photos, uh, trained them, and then I would generate the photos with the prompts, and I had this text file with the prompts, and I would do everything manually. And this b- quickly became way too much. So... But that's another constraint. Like, I was forced to, uh, code something up, uh, that would do that, and that was essentially making it into a real website, right?

Lex Fridman 58:12 So at first it was the Typeform and they uploaded through the Typeform an image.

Pieter Levels 58:15 It's a Stripe checkout Typeform. Yeah.

Lex Fridman 58:17 And then you were like, that image is downloaded. Did you write a script to export and download the image?

Pieter Levels 58:22 No, I just downloaded the images myself as a zip file.

Lex Fridman 58:23 Okay.

Pieter Levels 58:23 Unzip the zip file.

Lex Fridman 58:24 Literally. And you unzipped it one by...

Pieter Levels 58:26 Yeah, unzipped it. And then, yes, and then I... No, I'm... Because you know, do things, don't skill, Paul Graham says, right? So, uh, and then I would train it, and then I would email them the photos, I think from my personal email. Say, "Here's your, here's your avatars," you know. And they liked it, they were like, "Wow, it's amazing."

Lex Fridman 58:40 You emailed them with your personal email.

Pieter Levels 58:42 'Cause I didn't have an email address on this-

Lex Fridman 58:44 Yeah?

Pieter Levels 58:44 ... domain.

Lex Fridman 58:45 And this is like a hundred people.

Pieter Levels 58:47 Yeah. And then you know who signed up? Uh, like, man, I cannot say, but really famous people.

Lex Fridman 58:52 Mm-hmm.

Pieter Levels 58:52 Like really, really, like billionaires.

Lex Fridman 58:54 Mm-hmm.

Pieter Levels 58:54 Famous tech billionaires did it. And I was like, "Wow, this is crazy." And I sent... I was like so scared to mess with them, so I said, "Thanks so much for using my site."

Lex Fridman 59:02 Mm-hmm.

Pieter Levels 59:02 You know, and it's like, "Yeah, amazing app. Great work." So I was like, this is different than normal reaction, you know?

Lex Fridman 59:07 It's Bill Gates, isn't it?

Pieter Levels 59:10 Cannot say anything. (laughs)

Lex Fridman 59:12 (laughs) Just like shirtless pics.

Pieter Levels 59:13 GDPR, you know, like privacy.

Lex Fridman 59:15 Oh, right, right.

Pieter Levels 59:15 European regulation.

Lex Fridman 59:16 Right.

Pieter Levels 59:16 Cannot share anything.

Lex Fridman 59:18 (laughs)

Pieter Levels 59:18 But I was very, I was like, "Wow." And, uh... But this shows like, so you make something and then if it takes off very fast, you're like, it's validated, you know? You're like, here's something that people really want. But then also I thought, "This is hype. This is gonna die down very fast." And it did, 'cause it's too cheesy.

Lex Fridman 59:34 But you had to automate the whole thing. How'd you automate it? So, like, w- what's the AI component? Like how hard was that to figure out?

Pieter Levels 59:41 Okay, so that's actually in many ways the easiest thing, 'cause there is all these platforms already back then. There was platforms for, uh, fine tune stable diffusion. Um, like now I use Replicate. Back then I used different platforms. Um, which was funny, 'cause that platform, when this thing took off, I would tweet... 'Cause I tweet always like how much money these websites make. And then so the, the, he called vendor, right, the, the G- the platform that did the GPUs.

Lex Fridman 1:00:04 Mm-hmm.

Pieter Levels 1:00:05 They increased their price for training from $3 to $20 after they saw that I was making so much money. So immediately my profit is gone, 'cause I was selling them for $30. And I was in a slack with them, like saying, "What is this?" Like, "Can you just put it back to $3?" They say, "Yeah, maybe in the future. We're looking at it right now." I'm like, "What are you talking about?" Like, "You just took all my money," you know? And they're smart.

Lex Fridman 1:00:24 Well, they're not that smart, because like you also have a large, uh, platform, and a lot of people respect you, so you can literally come out and say that. But they're not sure-

Pieter Levels 1:00:33 Yeah, but I think it's like kind of dirty to cancel a company or something. I, I prefer just bringing my business elsewhere. But there was no elsewhere back then.

Lex Fridman 1:00:40 Right, r- right.

Pieter Levels 1:00:41 So I started talking to other, um, AI model ML platforms. So Replicate was one of those platforms. And I started DMing the CEO, say, "Can you please create..." like, it's called DreamBooth, this fine tuning of y- yourself. "Can you add this to your site? 'Cause I need this, 'cause I'm being price gouged." And he said, "No, because it takes too long to run. It takes half an hour to run, and we don't have the GPUs for it." I said, "Please, please, please." And then after a week he said-... we're doing it, we're launching this. And then this company became... It was, like, not very famous company. It became very famous with this stuff, because suddenly everybody was like, "Oh, we can build similar apps, like avatar apps." And everybody started buil- building avatar apps, and everybody started using Replicate for it. Um, and that was from these early DMs, with, like, the CEO, like Ben Fersch, very nice guy. And he was like... They never price gouged me, they never treated me bad. They've always been very nice. It's a very cool company. So you can run any ML model, any AI model, LLMs, you can run on here.

Lex Fridman 1:01:36 And you can scale.

Pieter Levels 1:01:37 Yes, they scale. Yeah, yeah, yeah. And, I mean, you can do now. You can click on a model and just run it already. It's, like, super easy. You log in with GitHub. Um-

Lex Fridman 1:01:45 That's great.

Pieter Levels 1:01:45 And by running it on the website, then you can automate with the API. You can make a website that runs the model.

Lex Fridman 1:01:50 Generate images, generate text, generate video, generate music, generate-

Pieter Levels 1:01:53 Video, like they do-

Lex Fridman 1:01:54 ... speech, fine-tune models.

Pieter Levels 1:01:55 They do anything, yeah. It's a very cool company.

Lex Fridman 1:01:58 Nice. And you're, like, growing with them, essentially. They grew because of you, because it's, like, a big use case.

Pieter Levels 1:02:03 Yeah. Like, the, the website even looks weird now. It, it started as, like, a machine learning platform.

Lex Fridman 1:02:08 Mm-hmm.

Pieter Levels 1:02:09 There was like... I didn't even understand what it did. It was just too, too ML, you know? Like, you would understand-

Lex Fridman 1:02:14 Yeah, yeah.

Pieter Levels 1:02:14 ... 'cause you're in the ML world. I wouldn't understand.

Lex Fridman 1:02:16 Now it's noob friendly.

Pieter Levels 1:02:17 Yeah, exactly.

Lex Fridman 1:02:19 (laughs)

Pieter Levels 1:02:19 And I didn't know how it worked. And, um, but I knew that they could probably do this. And they did it, they built the models, and now I use them for everything. And we trained, like, I think now, like, 36,000 models, 36,000 people already.

Lex Fridman 1:02:32 But is there some tricks to fine-tuning to, like, the collection of photos that are provided? Like, how do you-

Pieter Levels 1:02:38 Yes.

Lex Fridman 1:02:38 ... like... ?

Pieter Levels 1:02:38 Yes, man, there's so many hacks.

Lex Fridman 1:02:39 The hacks, yeah.

Pieter Levels 1:02:40 It's, like, 100 hacks to make it work.

Lex Fridman 1:02:41 What do... What's... (laughs) What are some interesting-

Pieter Levels 1:02:43 You can keep my secrets now and then.

Lex Fridman 1:02:44 Well, not, not the secrets, but the more, like, insights maybe about the human face and the human body. Like, what, what kind of stuff gets messed up a lot?

Pieter Levels 1:02:53 I think people... Well, man, it's a living... People don't know how they look. So, um-

Lex Fridman 1:02:58 (laughs) Yeah.

Pieter Levels 1:02:58 ... they generate photos of themselves, and then they say, "Ah, it doesn't look like me."

Lex Fridman 1:03:01 Yeah.

Pieter Levels 1:03:02 But it do-... Then I'll... You know, you can check the training photos. It does look like you-

Lex Fridman 1:03:06 Yeah.

Pieter Levels 1:03:06 ... but you don't know how you look. So there's a face dysmorphia-

Lex Fridman 1:03:10 Mm-hmm.

Pieter Levels 1:03:10 ... of yourself, that you, you have no idea how you look.

Lex Fridman 1:03:12 Yeah, that's hilarious. I mean, I've got to... One, one of the least pleasant activities in my existence is having to listen to my voice and look at my face.

Pieter Levels 1:03:20 Yeah.

Lex Fridman 1:03:20 So I get to, like, really, really have to sort of, uh, come into terms with the reality of how I look and how I sound.

Pieter Levels 1:03:29 And everybody, and, but-

Lex Fridman 1:03:30 People often don't, right?

Pieter Levels 1:03:32 Really?

Lex Fridman 1:03:32 Like, you, like you have a distorted view, perspective.

Pieter Levels 1:03:35 I know that, like, I would... If w- I would make a selfie how I think I look, that I- that's nice. Other people think that's not nice. But then they make a photo of me, and I'm like, "That's super ugly." But then they're like, "No, that's how you look, and you look nice," you know? So how other people see you is nice, so you need to ask other people to choose your photos.

Lex Fridman 1:03:53 Yeah, yeah.

Pieter Levels 1:03:53 You, you shouldn't choose them yourself, 'cause you don't know how you look.

Lex Fridman 1:03:56 Yeah, you don't know what makes you interesting, what makes you attractive or-

Pieter Levels 1:03:59 Yeah.

Lex Fridman 1:03:59 ... all this kind of stuff.

Pieter Levels 1:04:00 Exactly.

Lex Fridman 1:04:00 And a lot of us, this is a dark aspect of psychology, we focus on some small flaws.

Pieter Levels 1:04:05 Yeah.

Lex Fridman 1:04:06 This is why I hate plastic surgery, for example.

Pieter Levels 1:04:08 Yeah.

Lex Fridman 1:04:08 People try to remove the flaws, when the flaws are the thing that makes you interesting and attractive.

Pieter Levels 1:04:12 I learned from the hammerhead shark eyes this.

Lex Fridman 1:04:14 Mm-hmm.

Pieter Levels 1:04:15 There's stuff about you that looks ugly to you, and it's probably that, what makes you original, makes you nice, and people like it about you.

Lex Fridman 1:04:21 Yeah, yeah.

Pieter Levels 1:04:21 And it's not like, "Oh, my God." And people notice it. People notice your hammerhead eyes, you know? But it's like, "That's me. That's my face. So I'm... I lo- I love myself." And that's confidence, and confidence is attractive.

Lex Fridman 1:04:31 Yes.

Pieter Levels 1:04:32 Right?

Lex Fridman 1:04:32 Confidence is attractive. But yes, understanding what makes you beautiful. It's the breaking of symmetry makes you beautiful. It's the breaking of the, the average face makes you beautiful. All that.

Pieter Levels 1:04:42 Yeah, yeah. Yeah.

Lex Fridman 1:04:43 And obviously different for men and women of different ages, all this kind of stuff.

Pieter Levels 1:04:47 Yeah, 100%.

Lex Fridman 1:04:47 But the... U- underneath it all, the personality, all of that, when, when the face comes alive, that also is the thing that makes you beautiful.

Pieter Levels 1:04:56 Yeah.

Lex Fridman 1:04:56 But anyway, you have to figure all that out with AI.

Pieter Levels 1:04:58 Yeah. One thing that worked was like, um, people would upload full body photos of themselves-

Lex Fridman 1:05:03 Mm-hmm.

Pieter Levels 1:05:03 ... so I would crop the face, right? 'Cause then, then the model knew better that we're training mostly the face here. But then I started losing, uh, resemblance of the body, 'cause some people are skinny, some people are muscular, whatever. So you want to have that too. Uh, so now I mix full body photos in the training with face photos, face crops. Uh, and it's all automatic, like... Um, and I know that other people, they, they use, again, AI models to detect, like, what are the best photos in this training set, and then train on those. But it's all, it's all about training data, and that's with everything in AI. Like, how good your training data is, is in many ways more important than how many steps you train for, like how many months or whatever with the GPUs, like, um, the gold.

Lex Fridman 1:05:43 Do you have any guidelines for people of, like, how to get good data, how to give good data to fine-tune on?

Pieter Levels 1:05:48 Like, the photos should be diverse. So for example, if I m- only upload photos with a brown shirt or green shirt, the model will think that I'm training the green shirt. So the things that are the same every photo are the concepts that are trained. What you want is the- your face to be the concept that's trained, um, and everything else to be diverse, like different.

Lex Fridman 1:06:10 So diverse lighting as well,