Daniel Gross: Dreams and Determination

Hello, and welcome to The North Star. I'm your host, David Perell, the founder of North Star Media, and this is The North Star Podcast. This show is a deep dive into the stories, habits, ideas, strategies, and rituals that guide fulfilled people and create enormous success for them. And while the guests are diverse, they share profound similarities. They're guided by purpose, live with intense joy, learn passionately, and see the world with a unique lens. With each episode, we get to jump into their minds, soak up their hard-earned wisdom, and apply it to our lives. My guest today is Daniel Gross, a partner at Y Combinator, which is arguably the world's top startup studio. Daniel focuses on artificial intelligence at Y Combinator and recently founded AI Grant, a distributed artificial intelligence research lab. Daniel is 26 years old, but he's already had an incredible career. He was born in Jerusalem, Israel, and was accepted into the Y Combinator program in 2010, and at the time, Daniel was the youngest founder ever accepted. At Y Combinator, Daniel launched a company called Q, and after a couple years, Daniel sold the company to Apple, where he became a director at Apple for a couple of years. But then Daniel came back to Y Combinator to become a partner at the firm and focus on artificial intelligence.

What a career!

In this episode, Daniel talks about the experience of coming to America and starting a company with Y Combinator. He talks about the power of seeing life like a video game with levels and fast, continuous feedback. Daniel also shares lessons from John D. Rockefeller on business decision-making and company building. And finally, Daniel talks about Israel and shares his insights on why the country is so innovative. I hope you enjoy this episode. Daniel Gross, welcome to The North Star.

Thank you very much for having me.

So tell me a little bit about yourself growing up in Israel. How did you get here?

I was born and raised in Jerusalem, Israel. And I spent the majority of my life, 0 to nearly 18, growing up in Jerusalem, reading a lot about Silicon Valley, working on a lot of different pieces of software. Got into computer science when I was maybe 7 or 8 years old. My dad was a computer science professor. Got pulled into Silicon Valley when I applied to Y Combinator when I was 18. And so I was an Israeli, nearly soldier at the time. Everyone in Israel gets conscripted to the army, you know, waiting in line to to start my service, and I applied to Y Combinator, literally tethered my crappy laptop to my old Nokia phone from an Israeli settlement in the middle of nowhere.

Jeez.

And filled out an application, you know, took, you know, maybe 5, 6 minutes to submit, just waiting for that progress bar to submit with a terrible internet connection. And I remember at 2 o'clock in the morning, local Israel time, I was refreshing my Gmail constantly. I got that email from Y Combinator saying, that they'd like me to come out and interview. And in the back of my mind, I was thinking, fantastic, because at the time they paid half the ticket. And I was thinking I'd probably fail the interview, but I'd get a subsidized ticket to San Francisco.

First time coming to America?

First time coming out to the West Coast. My parents are originally American, and so I've been to New York quite a bit, never been out to the West Coast. I'd spent the majority of my teenage years reading about, you know, Microsoft and Google and even Fairchild Semiconductor, the old titans as well as the new. and was just excited to go see it. So I flew out to Silicon Valley, paid half the ticket myself, half paid by YC, and I sat down in the interview and I pitched my idea, and it went terribly. Went terribly, and I could kinda tell it was not getting any traction, but that was okay because my expectations were so low to begin with. I felt like I had little to lose. I walk out of Y Combinator and I'm walking down Pioneer Way, which is the street that YC still to this date holds, And I'm walking towards the Caltrain. It's about a 15-minute walk towards home, towards the train, and Paul Graham comes running after me, his, you know, hair flailing in the wind, and says, come back in. Pulls me back into the interview room and says, hey, so the idea you pitched is terrible, but you seem great. How about you come here, we'll fund you, and you just work on something else?

Wow.

And I paused for A couple of seconds, it felt like eternity, and I contemplated. On the one hand, I was quite excited about my idea. On the other hand, it's important in life, I think, when you realize, like, that a golden ticket is being offered to you, it's important to just take it.

You're like 18 years old at this point.

I think I was 17 at the time. I was just about to turn 18, yeah. And so I took it. I said, sure, I'll do it. Went back home, packed a suitcase, thinking again, I'd do a startup thing for 3 months, which is the length of the Y Combinator program, and then I'd come back home, it would probably fail, return to the Israeli army, Continue my service. Been living out of that same suitcase for the past 8 years or so. So been here pretty much ever since. Haven't gone back home. Very much focused on life here and work here. And so that's how I got here.

Before we go forward, I want to go backwards. So before the interview, you were asking me about some of the things that I was obsessed with as a kid and sort of like these just visceral obsessions where like you're, you're old enough to have them and sort of take agency, but maybe you get too young to realize that you're having them. What were those for you?

Give me an example for yourself for a minute.

So I mean, for me, it was math. I knew everywhere that you could go in San Francisco. I just— whether it was the topography of San Francisco, the subway maps of San Francisco, thinking about the way that fog rolls through the city. Like, I can still look at airplanes in the city and I know all the flight patterns in San Francisco. Yeah, sort of a weird thing, but I was just obsessed with it as a kid.

Very interesting. I guess for me, the ultimate thing I could never quite decode that I became fascinated with very early on in life, as far as I can remember, is really understanding the underlying motivations that drive people, because I think it's very counterintuitive. counterintuitive in the sense that people will tell you certain things. Like, if you ask someone, simple example, are you happy at work? You think they're just going to give you the answer of whether they're happy or not at work, but the more you think about it and the more you ask yourself that question, you realize they're kind of giving you the answer on the fly. They're kind of thinking right now, am I happy at work? What does happiness mean? And they actually answer a different question, which is whether that particular day was satisfying or not. So I think that's just an example of how counterintuitive the human mind is. And ever since I was young, I was very interested in just understanding more of that. And so I would spend hours— thankfully I had the guidance of my sister as a psychology professor— trying to use my own brain to kind of introspect on others. And this becomes— this is quite difficult to do when you're young because you're not fully developed. You don't really have spatial reasoning, as one simple example. But as you get older, your brain becomes a better mirror of adults, and you begin to realize, you know, why people act the way they do. And I've been fascinated by that for the longest time I can remember. Maybe the only other system I've been interested in mastering, just because it's more simple than people, thankfully, is computers.

Hmm.

I loved, you know, your obsession with maps. I loved breaking apart software, and initially I was very into reverse engineering software. Weirdly, I got into programming, very funny story, that I think is common amongst all of my friends, because I got into video games. And I really wanted to, um, I was playing Duke Nukem on my computer, and I really wanted to beat it. And it turns out one way to cheat in the video game is to learn where this program stores in memory certain values, like how much ammo you have and how much life you have. And if you do that, you can freeze those values, and so you have infinite ammo and infinite life.

Wow.

This is how video game cheats work. So that's how I got into coding. It was all because I wanted to reverse engineer a video game in order to win.

So, so I was planning on talking about this later, but it's really interesting, and it's something that you wrote about a couple weeks ago with English subtitles as a way to learn languages. And you can watch a movie as a foreigner, and you're having fun watching a movie, but you're learning all the same. And it's funny because when we think of school, At least me, I think of just total boredom. But there's ways to learn where it's super fun and you're learning. Talk a bit about that. I think every gamification of learning, I guess, might be the best way to put it.

Yeah, and I think gamification is, oh man, like an old ex-girlfriend. It carries a lot of baggage with it because it's been around for a while and it means a lot of different things. And I think in many people's minds it evokes thoughts of, you know, Zynga and FarmVille. But really, to me, what's so fun, right? That's, that was the point of it. Everything in life ultimately is some form of a video game, and when things are great, when you're in flow, really what's happening is you're at the proximal or optimal state of learning, where you're being challenged enough so that it's interesting and there's novelty, but you're not being challenged too much to the point where it's draining.

Right.

And really what a bad day is, it's a video game that was not well designed. If you think about it, in the sense that you just got too much negative reinforcement and that wasn't fun.

Hmm.

All humans, regardless, I think, of their IQ, ambition, everyone just wants to— everyone is novelty-seeking, and they're trying to arrange the world in their favor. That's what you do as you walk around the world, whether you're getting lunch, going to meet someone, or going to watch a movie. You're trying to arrange the world in your favor, and you're ultimately always trying to seek novelty and also approval of others. We could get to that later. And so I think, you know, video games are just an interesting colloquial way of dissecting what's been going on forever, which is people have been trying to optimize the rate of learning that they're at so that things are equal parts, you know, interesting, which I think actually means challenging, and fun. So you have to win a certain amount. Win too much and it's boring.

Right.

You know, win too little and it's too challenging. So the question in life is is, you know, so kind of Nintendo video games are literally designed to be perfect like this. You know, Halo, as an example, will match you up only with people that are at your level. They'll assess your level and they'll match you up with people at similar levels. Of course, life doesn't do this for you. And so the question is, how do you create your own life such that you're always in this optimal state of learning? And I think a lot of this comes to, you know, not putting too much on your plate so that you feel like you're failing, but putting enough on your plate so that you feel like life is interesting.

Mm-hmm.

And so in a nutshell, yeah, I mean, it's, it's a little nerdy and weird to say that everything's a video game, but I just think it's a helpful metaphor for us to understand what our minds are looking for, which is the right level of novelty at all times.

Talk about approval of others. You mentioned it briefly. Let's go down that rabbit hole.

Yeah, I mean, I think people don't like— it's, it's always funny to ask people, what are you motivated by? Because people feel like they have to come up with a very original answer, you know, and, and there's certain things that you're not allowed to say that everyone's thinking when someone asks you, what are you motivated by?

Yeah.

Because I think everyone's thinking, well, you know, it would be nice to have a slightly larger apartment, and it would be nice to be able to get a fancy dinner once in a while, but I can't say I'm motivated by money. It also, you know, another thing that I think people feel but don't even— it took me a long time to learn how to express— is the approval of other people. Boy, it would be nice to have it, and it feels, it feels wrong to to suggest that you'd want that, but I think ultimately it's very healthy to say that I, as a human, want the approval of others. Now, it's important to differentiate looking for the approval of the masses, which is, I think, not healthy, and that's kind of the celebrity effect where you want millions of people to think you're great. That's maybe what people think Donald Trump is after, Right. versus the appreciation of your peers, of your friends, and that is, I think, very healthy to do because that is merely how you find out if you're good or not.

Totally.

To go back to our video game analogy, that's how you know where you rank on the leaderboard. If you're doing something and all of your friends are telling you it's bad, then that's really helpful information for you to not do it anymore. And if they think it's good, it's really helpful to know that you should do it again. And so it's totally okay to say, well, I care a lot about not what the masses think of me, but what the select few think of me. And I think that is— I think that drives a lot, if not all, people. Wouldn't say all. It's, you know, humans are thankfully so different. There's always exceptions to the rule, but I think many of us are driven by getting the approval of our peers around us and getting them to think that we're okay. And I think that's very healthy and not something to be ashamed of at all. And, and so yeah, so, so that's why I think basically life is, is, you know, you kind of navigating the world trying to get what you want. And I do think that one very core component is getting other people to think that You're great.

Love that. Let's talk a bit about what happens after you arrive in America. You go to YC, your company gets acquired by Apple. We'll talk a bit about AI, and then we'll jump into feedback loops.

Okay, sounds good. So tumultuous years. The sad truth is, you know, the years of 2010 to 2013 where I was running my company, 2013 to 2017 when I was at Apple, were—

You're what, 20 years old?

When we got acquired, yeah, I was 22, I think. 21, 22. Um, there was a lot going on, and so the brain has compressed a lot of those memories, not a lot of sleep. But, um, uh, so after Y Combinator, I, um, the YC, uh, path was kind of tumultuous unto itself. I, um, had to scrap my— I lost my co-founder, and I had to kind of redo my entire idea yet again about 48 hours before Demo Day, this, this, uh, you know, massive event at the end of Y Combinator where the best investors in the world come by and listen to everyone's pitches. Um, and about 48 hours before, the project I was working on, for various reasons, I had to totally scrap and cancel. I was getting sued by a large company.

Oh man.

Um, so I had to come up with a brand new idea, lose the co-founder, and basically kind of Superman a brand new project, which ironically, ironically, was always the thing I wanted to build. And this was one key lesson in life I hold dearly to this day, is I only had the ambition to kind of build the moonshot idea I'd always wanted to do when I was put under the gun.

Oh.

I remember going to Paul Graham's house, this is 48 hours before Demo Day, and saying, well, I gotta think of a brand new idea, what should I do? And he says, look, you have 3 options. You can pitch your old idea and not tell everyone that you just got sued by a large company. You can come up with a brand new concept and build it, and build a prototype in the next 48 hours and try to demo it, or you can defer Demo Day and just do something later on. And I said, well, I wanna do the second, and he asked me what ideas I had, and, uh, we were walking around his house. This is— the true irony here is we're walking around his house in Palo Alto, and in a true foreshadowing of what was going to happen in the future, I have my only encounter ever with Steve Jobs, because he almost runs me over in his car.

What?!

Uh, yeah. Um, and so I described this product that I'd always wanted to build, but I was afraid to, uh, and it was, uh, basically a search engine for all of your personal content. So you could search the web quite easily with Google, but You know, increasingly, we all have multiple Google Apps accounts, and Slack, and Gmail, and Dropbox, and just like navigating to different files is a pain. And so I wanted to fix that. And, um, I was always afraid to build a search engine. It felt like something people at Google do that you don't do. And, um, I learned that you can pretty much do anything you want in the world, as long as you dissect the problem and turn it into an interesting video game, if you will.

Mm-hmm.

Uh, and so I just started working on that in 48 hours, built a really stupid prototype myself, got up on stage, demoed it, raised a Series A from Sequoia, I mean, brought on a co-founder who became a very good friend of mine, um, who was working at Google at the time, raised a Series B again from Sequoia, grew the company, uh, ended up getting acquired by Apple in 2013.

Wow.

To work on Spotlight, uh, which is basically the same thing on your phone. Ported a lot of our technology over right to the iPhone and, uh, OS X, where it runs to this day. And, um, also started working on some broader machine learning and AI projects at Apple across the entire organization.

When I go on my iPhone, what things that I'm using is the code that you ran?

Ironically, it's an eclectic mix of things, so it's not like I can point to one particular thing. Probably the most notable— a few notable examples are when you swipe down and go into Spotlight, the fact that it predicts what apps you want to open, the fact that it searches apps significantly better than it did before, you know, the fact that it just kind of works better is one. Occasionally you'll get a phone call and you'll get a caller ID match even though the person's not in your address book.

Yeah, yeah, yeah.

That's because there's software running on your iPhone mail client that's trying to understand when people send you their phone number and they say, hey, I'm free for a call, here's my number. It automatically adds that to your address book. It also does the same exact thing for events, tries to automatically detect events using machine learning, put them on your calendar. And so there's basically this proactive assistant that is constantly working on your phone, filing away, doing things that a normal assistant would do if they had full access to your phone. And that is entirely all of our team. Pretty much all of them are still there at Apple working away at this. So They get, you know, almost entirely all of the credit for it.

Wow. So in terms of artificial intelligence, you come back to Y Combinator. Tell me a little bit about your pattern of learning about artificial intelligence. I'm interested sort of in the models that you had as you were learning that really helped you understand AI, sort of like an AI for dummies here.

That makes sense. I mean, one thing— one, if I reframe your question a little bit, if you don't know much about AI today and you're trying to get a mental model for it, One thing that is helpful is to really think of it as trying to convince an autistic savant to do different things.

Okay.

In the sense that it's incredibly brilliant at operations. You know, you'll read these stories about Daniel Tammet, you know, that autistic savant that can just, you know, memorize all the— or not all, there aren't— there's no such thing as all the digits in pi, but some of the digits in pi. Do incredible math in his mind, but, you know, very socially hampered in many ways. Actually, Daniel is one of the most productive people in this category, but occasionally these savants, when they're over-indexed on IQ, kind of lose out on EQ, lose out on other things.

Yeah.

And AI is like that. It's even an autistic child, it's even more hampered, but it's really good at doing simple things. And this is an important construct because if you internalize it, you'll realize the mistake most people are making is they're walking around and they're waiting for computers to do things that humans are doing. A lot of the things we do are really hard for computers to do, but computers can can do other things that are really amazing. So it's a little bit less like you're gonna get a household robot that's gonna have the dexterity of your fingers, and it's gonna be able to do your laundry, and a little bit more like you're gonna write software that solves math theorems really well, or you're gonna write software that can unders— that can hear at a superhuman level. In general, another mental model I think I developed, not just with AI but more broadly, is when you're trying to predict the future, the mistake that you make is you take the current reality and you project it forward.

Yeah.

And the challenge— and humans have been doing this forever, right? So like the first TV shows are cameras pointed at radio shows. Or when people build mobile apps, you know, circa 2007, what they do is they take the large tech company Google and they build it on mobile, like mobile web search. Turns out mobile web search is Google. It turns out mobile banking is Chase. It turns out the real kind of mobile app is Uber, is Instacart.

Mm-hmm.

The real TV show is Game of Thrones. You don't think of Game of Thrones when it's, you know, 1910. And, you know, TV screens are barely coming out of the lab. The mistake is, in general, a flawed mental model is the use of analogies. What's going on here, by the way, is that analogies are mentally very satisfying to think of. Like, when they click in your head, when you think of an analogy, it feels good, and it feels good to describe it to others.

Yeah.

And so it permeates through society.

Right.

There's a great area of social science called the availability cascade, pioneered by Timur Kuran, who studies this, who basically posits that a lot of the beliefs that we have in society are basically these things that kind of mentally, they're not correct in an ultimate truth sense, but they kind of click when you hear about them.

Yeah, yeah.

And so, 2 things are going on. One is, it feels good to learn, and second is, it feels good to teach that to other people, because that's another way of kind of saying that you're smart. I'll give you a very succinct example to make this more practical. GMOs are bad for you. Are they? Most people, I would bet, discussing this don't really know the science, but it feels right that the evil people Monsanto that are engineering grains are causing you harm. Or another one is MSG causes headaches. So someone tells you, you know how when you go to a Chinese restaurant you get some headaches?

Or fat makes you fat.

Or fat makes you fat. It just feels correct when you hear it, and so you want to tell other people about it so that you feel like you can teach them. And again, things permeate through society. So I think at the end of the day, a lot of these truths that you have that are built from analogies are actually wrong, but it felt right. So you think about it. And so AI, again, going back to your original question, don't like— don't try to like superimpose the current futuristic thinking. I think what's more likely to happen is something weird that you didn't expect. Like for a while before crypto got big, we were all thinking that the next wave in the internet was of course going to be AR and VR. And maybe that will happen in time. But then again, we got whacked out of nowhere with crypto, which is like weird, not quite what you expected. And so with AI, I would, I would be more interested in, in people brainstorming like, what is the Game of Thrones equivalent? If we're listening to radio shows and we're trying to figure out TV shows, what is Game of Thrones for AI?

Do you have any ideas on this?

I think it's going to be things like solving math theorems. I think it's going to be things like making music that is significantly more pleasurable than human music to listen to. It's gonna be stuff that's weird, and it's going to be stuff that is often overlooked at first. This is always what happens. I mean, you could go back and look at the great video of Steve Ballmer laughing at the iPhone. Everything is perceived as a— or Tesla Roadster— as a toy at the beginning.

Right.

And it becomes mainstream. It's one of the reasons why startups always win, is they can apply attention to the things that large organizations are coining as kind of under or overrated or not important to look at. And so I suspect when you think of like what will get really big in AI, it's going to be a weird thing like, you know, like music, like art, where you can throw it away and say, oh, there's no business model there. It's just some stupid project at Berkeley or in India. And that, that's going to get really big.

Can you talk a little bit about stuff that you're doing with AI specifically at Y Combinator?

I think one of the main constrictions we have on all scientific progress to date is that we don't have a diverse enough set of brains working on it. And I don't mean this in your classic diversity sense, although that would be great to fix as well. I mean this in just like volume of different ideas coming through the system is too small. The extreme example of which is I think like 30 to 50%, depending on the year, of research papers that are published on arXiv— this is a website that publishes machine learning papers— are all published by Google Brain. And Google's amazing, but When you work at Google, first of all, you only get to working at Google if you look and operate a particular way. And you get indoctrinated into the culture, and that has a certain specific way of thinking. And to our earlier point about seeking the approval of others, you probably want to do what your boss says is good because you want his approval. And so that creates groupthink, which is often good, but that's only one type of group. And so I think we're actually stymied by the fact that there's not enough different weird, almost stupid and childish ideas being explored. And so my work right now is to kind of increase the richness of the petri dish, if you will, increase the kind of genetic diversity of the thinking. And I'm trying to do that both with startups and in research. So with startups, I started this thing called YCAI. And what it tries to do is it tries to level the playing field in starting an AI project, because if you start an AI project today at Google, you actually, you actually are significantly more advantaged than a startup. In 3 particular ways. One, you have access to a density of expertise and talent that is locked up at Google.

Hmm.

They basically have a lock on many of the great researchers in the world, and you need great researchers today. It's not a commodity where you can kind of hack stuff together. Second is you can spend a lot of money upfront on compute, and, and that's somewhat equivalent to what the web used to be like before AWS, where you were able— where you just had to go rack servers in a data center somewhere, and it cost half half a million dollars to put up a website. AI is like that today. And then the third thing is probably the most important, is you have proprietary datasets that, that basically are the currency for intelligence in AI today. Computers are not able to do this thing that humans do very easily, which is what's called transfer learning. So when I go play a video game, say like Pong, I can walk up to that video game and I can play it for the first time and master it.

Hmm.

I don't need to train a billion times. And the reason why is I have a sense of what gravity is, and I have a sense of what objects do when they hit each other. And so I know as the little ball is heading towards the paddle in Pong, I know that it's gonna bounce back. Computers can't do that cross-domain analogy.

Hmm.

And so as a result, we need a lot of domain-specific data. Like if you want to build an AI that's a better radiologist, you need a lot of very specific radiology data. You can't give it other X-ray data. And so those are those 3 moats that I described that companies have. And what I'm trying to do at YC is give those 3 moats to startups, basically. So we started this thing called YC AI, a vertical of the program dedicated to machine learning and AI companies. And startups who are in it basically get those 3 perks. So talent, they can basically book office hours with people who work at OpenAI.

Mm-hmm.

So they get access to the brains, basically. In terms of compute, we give them an enormous, like, high 6-figure amount of dollars in terms of GPU resources they can use in addition to the YC investment. We even work with a company in San Jose that just racks GPUs for our startups to use.

Wow.

And this turns out to be really important. Like, if you're trying to build a better version of Siri, you have to spend half a million bucks on training your models first. And the third thing that we try to give our startups is datasets. So working with different companies to provide startups with datasets. So that's what I'm trying to do on the company formation front.

Can you give us just a specific example of a company that's been able to leverage the YC resources and walk me through what they wouldn't have been able to do before and what they can do now? Yeah, like a specific example.

Very, very, very specific and awesome examples. This company called AssemblyAI, and they're building a better version of Nuance. or a better version of Google's speech recognition API.

Wow.

So they're going— I mean, they're, they're, they're David chasing the Goliath. In many ways, this has worked in the past. Stripe is an example of company— of a company that basically did that in the payments world. You know, they went head-on against PayPal, provided a simpler interface, something that developers loved, and ended up becoming tremendously valuable. And so this person is basically doing that for speech recognition. And if you take the view of the world that voice interfaces are kind of the future of how we're going to interact with computers, this could be as big as the keyboard and mouse was. Nuance alone, which is an independent company that does this today, is, you know, I think worth a couple of billion dollars. And so AssemblyAI was basically able to get started and train their models because of the credits they provided with us and because we reduced the activation energy sufficiently. It turns out that the venture capital community kind of won't fund you until you have something interesting. And in order to have something interesting, you have to spend money. So you have this bootstrapping problem. And so they're an example of a company that was able to bypass it. The other, you know, the most common thing, weirdly, the companies get out of the program is maybe a 4th benefit, which is they just get density and camaraderie with other companies working on AI.

Mm-hmm.

And it turns out that when you're in a particular domain, when you're going through a shared experience together with people, and this is true for YC as a whole, but particular people in particular domains, if your sense of suffering is shared, if you're both having the same challenges, You build true bonding.

Right.

And so I think they all benefit from that in the same way that people who, you know, run a marathon together benefit from, from the sense of bonding. When I meet another runner, I kind of know we both go through something hard together. And so, you know, suffering in many ways, or challenge, creates relationships, really strong relationships.

Totally, totally.

And so I think this kind of subreddit almost for YC creates stronger bonds than would have been possible. Than previously.

I gotta ask, because you talked about datasets, I'm really curious to understand how your understanding of artificial intelligence and how computers operate impact the way that you live your life. How would you think about getting more datasets, so to speak, in terms of your own lived experience?

Well, I'll answer your question. You let me know if I answered the right one. Now, if you get dismayed at any step, just realize that what's happened here is the video game has slightly become too hard. So if you spend 3 hours trying to like download a website to your computer with a script and it's not working, you're gonna get drained, and you're gonna get tempted to give up on the entire project. Have no fear. All you need to do is pick a smaller step. Turns out you, you're trying to chew too much at once. The other useful thing you can do if you're really struggling is create and write down the version of yourself that doesn't do this. So create a negative reality. And so write a little— I would just write like a plan. So, you know, you have your little project that you want to build in 6 months, and You could think about it in your mind, get really excited about it, but I would also write down the version of you that doesn't build this project. You know, what do you spend that 6 months doing? And if you can come up with a better idea of how you're gonna look and how you're gonna spend your weekends, maybe you should just go pursue that. But if you can't, use— write it down and use that negative reinforcement in addition to the positive reinforcement to propel yourself forward.

Okay, so one thing I struggle with is, and a lot of people that I know too, is this trade-off between short-term learning and long-term learning. So Say I want to learn Python, which I do. One year from now, I want to have learned Python, but it's not as much fun, and it's hard to do on a day-to-day basis, and I would almost rather default to reading.

But over—

if I make reading a habit, then I might not learn Python. So I'm wondering how you balance the, the potential tension between short-term learning and long-term.

Yeah, I think I think the answer to that is you've got to build yourself a plan. Or to say it in a slightly more interesting way, because that sounds boring, you've got to build a video game for yourself.

Mm-hmm.

And, and the way I would go about doing that is I think you basically need 2 things. One is you need a vision of where you want to get to, which you have. I want to learn Python. I want— and I would think long and hard about the fantasy there, because that's gonna— what's gonna get you going through the hard times. Visualize yourself knowing Python and really think about what that means, because even as I say it now, it sounds a little vague just to know it. So maybe it's not I'm gonna learn Python, but like I'm going to build a thing, and I would get really excited about a particular thing.

Mm-hmm.

I know this happens to me, and some people claim that they're not ideas people. I think that's crap. I think everyone loves ideas. Just think about something you would love to create.

What's an example of something that you've done in your life In terms of thinking of something, making a visualization, going to learn, and making that process happen?

I mean, my problem is I do this too often. So I have a Google Doc that I dump 5 ideas a day into, and occasionally I'll go after and start building something. And I'll confess, it's a weakness of mine. Occasionally I'll start doing that even when it's not the topmost priority, just because I derive so much joy from it. Ideas infect me in a very aggressive way. I'll give you a very simple example of something that's infected me of late, but it'd be really neat if Goodreads was much better. The fact that you read a book today and it's basically a single-player experience sucks, and I think you could make reading significantly more interactive if I just knew what highlights my friends were doing. And so, okay, that's, that's the grand vision. And so I'll start thinking to myself, what's a practical thing I could build that would be interesting? And I think, well, maybe I should build a kind of a script or an app where I could just import all of my friends' reading progress and we could have like a dynamic leaderboard where we're seeing how much everyone's reading every day. And am I reading slightly more, slightly less? Because Kindle has all that data. And then I'll start going out online and I'll start trying to see how can I interface with the Kindle API? And it turns out it's proprietary, but it turns out someone on GitHub has like published a thing that interacts with it a little bit. It kind of makes believe it's its own little Kindle. And so I start building that and then I start asking my friends for their, you know, Kindle passwords and we start piecing something together. And of course, the challenge with these projects is the 80/20 rule. It's really easy to do everything I'm describing, really hard to ship it. You know, 80% of the effort is the last 20% of the project. But that's an example of something that I will just get affixed with. And a good biomarker for when an idea has truly infected you is when you're thinking about it in the shower.

Mm-hmm.

And so I have a lot of those where I just cannot stop thinking about something. And it's occasionally— it's an impediment. But going back to your question, I would think less about Python and think more about Think about the world, think about little annoyances that you have, and just think about how you'd fix them. Like another small annoyance I had today, trying to dial into a conference line, and I have to constantly like memorize the PIN code. So like you press the number, if Google like doesn't take you to Google Maps and actually manages to start the call, you have to like go back to the calendar app, like keep the number in your head, paste it in. This should just like be a solved problem. So maybe you should build an app that solves that problem and just puts in a one-tap thing in every single calendar event. have. Great. So there's thousands of these like little neat ideas, and some of them could be great businesses. Some of them are just gonna be neat ideas. If you just want to learn, pick any. Doesn't matter if it can't turn into a business. Get excited by building it. So that would be the first thing I would do is I would crystallize the vision. That's part one of making the video game. The second part of making the video game is reverse engineer the vision into a plan, because the vision is It's like saying, I'm really excited about level 10 Mario. Great, but you don't even know how to grasp the controller.

Right.

So build level 1 for yourself, and that probably means thinking about literally the, like, the dumbest thing you can do to get to learn to code. Commit to Googling around different ways to learn to write scripts— Python, Ruby, what have you. Just say, I'm gonna do that. That's it. That's really easy to do. And plan that for tomorrow. Go ahead, do that the next day. Reward yourself by doing it, like mark that off your checkbox and treat yourself to, you know, a coffee, whatever you want.

Tastes like latte.

Latte, whatever it is that treat is for you. Feed that system, eat some sugar, reward yourself. Think about the next small step, you know, which in this particular case might be try to build something that— let's imagine you are trying to build, you know, a better version of Goodreads. Try to build something that can connect to any website. Doesn't even have to be interfacing with the Kindle website, is actually quite hard. Just any website, just download any web page.

Right.

Do that. That'll take you a couple hours. Keep the vision in mind. Now, if you get dismayed at any step, just realize that what's happened here is the video game has slightly become too hard. So if you spend 3 hours trying to like download a website to your computer with a script and it's not working, You're gonna get drained, and you're gonna get tempted to give up on the entire project. Have no fear. All you need to do is pick a smaller step. Turns out you, you're trying to chew too much at once. So maybe don't try to download a website. Just dial it back down to like writing some code that prints out some words. Reward yourself for that, close the laptop, and leave.

Right.

So I think it's— I think the, this, the other mistake people make is when things go south, they tank on the whole thing. But it's all about saying, okay, that was too much. Let me start with a smaller thing. Another analogy, if you do this, is weightlifting. Well, you know, if I, if I go try and bench 300 pounds, it's not gonna work. And so what I'm gonna do is— it's not that I'm gonna leave the gym at that point. It's, oh, I'm gonna try to bench 45 pounds and then slowly build up.

One more thing that I would add to this, directly inspired by your work, is to cultivate an environment that's gonna help you.

Yeah.

And let's talk a bit about that, and then I want to jump into your knowledge of Israel. And what it is about Israel and the Israeli environment that has spawned people like you and some really phenomenal inventions.

Yeah, environment, that's true, that if you can, environments are the ultimate cheat code for all of these things. They will, basically what they're gonna do is all the things I'm describing automatically for you. And so what I mean by that is, if you can surround yourself in an environment of other people that are building, it's gonna be much easier for you to build stuff, because again, you're gonna wanna win the approval of your peers, your peers will be building, and so you'll be like directed into doing that.

You know, it's funny because it reminds me of that Paul Graham essay about cities and what cities whisper. And like Boston sort of whispers about academia, San Francisco is like, be, you know, make more money, LA sort of be more famous, New York is, you know, make more money too. But cities whisper different things, and ever since that essay, I've been thinking a lot about what city do I live in and how is that gonna push me. Like New York is like this 20-mile-per-hour wind at your back in terms of just doing stuff. Anyway, back to what you were saying.

I would agree with that framework, but a more optimistic interpretation of that is you do not have to optimize for a city as much as you have to optimize for the right small group.

Hmm, interesting.

And I think that too little has been written about the biographies of small groups as opposed to the biographies of great people. Yeah, a lot of the advancement in society, you know, Xerox PARC is a famous example of a small group But I actually think that a lot of it, whether you look even at the single operators like Rockefeller, Einstein, Marie Curie, it's actually the small peers that we're surrounded by. In Einstein's case, his wife, other professors he was writing to, that really motivated them. Again, we're all seeking approval of our peers, and we're all going to be kind of the average of people we spend time with. So it is true that the ultimate, penultimate way to become like an engineer is to go live in the Mission in San Francisco, because everyone else is an engineer. I think if you just— even if you're living in Nairobi and you surround yourself with other 4 or 5 engineers, that could be enough.

And now you can actually even do it on Twitter just by sort of cultivating your streams. You build your own little echo chamber.

So that is, of course, the ultimate question for us, is, is it possible to create these small group bonds that set yourself up with a great environment online? And I hope the answer is yes, because if it is, the world is set to accelerate to a whole different degree. Because it used to be the case that if you were born the Ramanujan in India or the Einstein Einstein as a patent clerk in Bern, you were stuck until you physically moved atoms around you to kind of, almost by luck, build yourself, you know, an environment kingdom. But if it's the case that you can do this digitally, that's really exciting, because then you find the subreddit, and you subscribe to it, and you're ready, set, ready to go. And it's all— the key to this question is, are the relationships that you build online strong enough such that you're motivated to make those people think you're great? That's the question. And so, you know, it's very clear that if we were neighbors or if we were roommates, I'd be very interested in you, and I'd be very interested in what you thought of me. And if you thought I was doing something bad, I would consider myself— well, I kind of need to check what I was doing.

Mm-hmm.

And so there's a lot going on in the mindset of who you physically surround yourself by, and the question is, do you engage that same software in your head if you're surrounded by those people on the internet. Sometimes you hear about people that develop true friendships on the internet, especially in the old IRC days where it was kind of weird to be on the internet. But I also have, you know, to the other extreme, I really don't care what most people on Reddit are doing. You know, I don't, I don't know who they are.

Mm-hmm.

And so there's, there's a very particular set of software that gets installed in your head when there's people physically around you, and I hope that That some online communities can replicate all of that. But, you know, in terms of what you should do if you're listening to this podcast and you're not somewhere, you know, where you can easily move into a great environment, first of all, congrats on finding this podcast. I think that's the first step. No, I seriously think surround yourself in a digital environment that is even— it's asymmetrical, so it's not great. Like, we are talking to people that are not talking back to us, but that's still one great first step. You can surround yourself with books of great people. You can surround yourself with other podcasts. That, that's one good first step. The second step is you try to find a digital community that you could talk to. And so I hope you can find some type of WhatsApp group or Telegram group or, or Reddit where there's people that you build relationships with such that you can really leverage them to kind of level yourself up. The third step in terms of optimizing your environment is physically locating yourself near those people or the people of that mindset. So in your earlier example, people looking to write Python. And then the final example, if you really can afford it, is go to a city that whispers a certain way. My hope is that you can do this with small groups, not with cities, because then it is much more scalable, right? And so I agree with you that the ultimate panacea is building your environment, and there are kind of different levels of environment here that you could build.

Talk about your experience in Israel, because Israel has like this very special environment, some of it actually for unfortunate reasons, but It's really interesting to think about sort of the security reasons, the water, sort of all the water innovation that has kind of come out of Israel, and to sort of think through the culture, the shared bond that I think Israelis have because almost everybody serves in the military, and also the, the constraints— not having oil and having a lack of rainfall and having to be really precise with water usage.

The question you wonder is, of course, are those things sadly linked? And I think the answer is yes. That is to say, I think Israel's greatness stems a lot from the fact that it had to be great. Forcing functions are a wonderful thing. One of the most potent things that happens at YC is people race really quickly towards Demo Day.

Mm-hmm.

One of the great things I saw inside Apple is people race incredibly aggressively towards WWDC, Apple's annual developer conference. And so you really need a forcing function in order to succeed in life. Which is maybe something to include in your— how do you get better at Python?

It's funny because, so my best friend after college worked at Yelp, and his team did like 30% of revenue on the last day of the month, every single month.

Yeah, and I see for YC and AI grant, most of the applications come in moments before the deadline. So to your point, I mean, we should go back a little bit. How do you get better at Python?

Well, um, and Israel too.

You need to take your— you need, you need to build a forcing function, and a way to build a forcing function is artificially is to tell a lot of friends you're gonna do something by a particular date. Friends that you care about. Don't, like, don't tell it to someone who you're not really gonna care what you think. Find people who you respect, tell them you're gonna do something by a certain date, and in all honesty, they're not gonna care. They're gonna be like, why are you telling me this? Suddenly you're gonna find yourself, without really thinking about it, just working really hard to get that goal done. At least in my mind, I build this narrative in my head that they are gonna care and I don't want to let them down. So build a forcing function for yourself. Back to Israel, it's harsh, and I think that harshness creates the greatness. So there's a sense of camaraderie and bonding, again, I think, like in every community where there's a shared sense of suffering. And this happens both on the Israeli and the Palestinian side. In many ways, on the Palestinian side to a greater degree, because I think the quality of life there is sadly much lower.

Mm-hmm.

And so, yeah, when you meet— when I meet another Israeli who grew up in Israel during the era I grew up, especially during the Second Intifada, or even in the first. And they, and they, you know, they say, yeah, I experienced that, basically.

What is the Intifada?

Intifada is a period of years, happened twice, where they're basically— it describes a period of years where there's a lot of unrest in both the West Bank and Gaza Strip. And so there's a lot of back and forth. And so, you know, circa 2005, there was like 2 suicide bombings a day in Israel. And the situation's quieted down quite a bit. And I think that humans are never really grateful, weirdly, for this. We'll be grateful for this era if, you know, heaven forbid, we get a 3rd intifada. But things are actually really quiet now compared to the era I was growing up, where, you know, you'd be sitting in traffic behind a bus and you'd be wondering if you should change lanes, if it's going to blow up. You know, and I mean, and to your point, incredible amount of technological innovation. Buses developed these, like, really complicated X-ray machines where you got screened as you boarded the bus to make sure you weren't carrying a bomb. But if you did, The bus could kind of cut you off so that if you detonate it, no one else would get hurt. And that stuff got built really rapidly. Or Iron Dome, great example of a system Israel uses to this very day to shoot down rockets that come over from the other side. It's another fascinating story of a forcing function. So the, you know, the U.S. has this thing called a Patriot missile, which does a similar thing. It costs, I think, hundreds of thousands of dollars per missile. So it's not economical. to use it to shoot down mortars that cost $100.

Mm-hmm.

Because they just fire a bevy of them, and then you have to blow millions of dollars on shooting them down. So the United States ends up working on this thing called Nautilus, which is literally out of Star Wars. It is a massive laser cannon that is going to use high-powered electro-microwave waves to shoot down the rockets. It works on it forever. No forcing function. Spends billions of dollars on it. Doesn't go anywhere. Israel, in the meantime, is like, hey, we're getting nailed with these rockets. And I used to briefly live down one of these settlements that was plagued by— it wasn't a settlement, it was a proper city, and it's a terrible quality of life. Basically, a siren goes off and you have 5 seconds to duck for cover before maybe dying. And so you're walking around town, even when the siren's not going off, you're constantly wondering, where will I duck to? Where will I duck to? You know, if the siren goes off. Weirdly, it's actually quite calming when the siren finally does go off and you have to race to cover. So what Israel does is they get together a crack team and they say, look, you guys have like basically no budget and a year and a half to build Iron Dome, and you're going to have to shoot down rockets using whatever you can figure out, you know, duct tape and chewing gum. And they, and they get it done. So, so that story really should give you a sense of, I think, why Israel is great, is it's because it has this pressure to be great, where that doesn't exist in actually countries aplenty like the United States. Why should the United States be good at desal? Why should anyone be good at desal? You have plenty of water. Turns out Israel doesn't have plenty of water, so it gets really good at it. And so scarcity creates greatness. This is another interpretation of why many people believe the northern part of the United States is significantly more developed in terms of economic infrastructure than, say, Mexico. Why is it that the quality of life in San Diego and Tijuana is so dramatically different despite them being miles away from each other? One hypothesized answer in this book called Why Nations Fail, one hypothesized answer is the culture of adversity and scarcity that you had in the northern United States creates this Puritan work ethic, creates solid banking infrastructure, creates degrees of trust that you don't develop in highly corrupt societies, which develop because you have the curse of the plenty. So people go to Mexico to chase gold. And it's warm and it's sunny and there's plenty of crops. New York, by, you know, by 17th, 18th century standards, sucks. It's basically the leftover pickings of the New World. But that created significantly better economic infrastructure because that was the only way to survive. And so I think in many ways that's why Israel's great. There's a lot of other theories people like to hypothesize, you know, the particular Ashkenazi Jew focus on scholarly leadership and you know, the Middle Eastern kind of culture of do whatever it takes to get the job done. But I think a lot of it is basically a nation that has had to succeed. One particular feeder for this that's very interesting to look at, one final point on this, is the Israeli army. Everyone in Israel, literally everyone, gets psychometric and physical screening when they're 16. And they basically pick the best in the country, and they put them in various units, you know, be it the Air Force or the Israeli intelligence, It turns out, if you're able to do that, and you put all the smart, ambitious people together, you get other great dividends, second-order effects. And so there's a lot of startups that come out of this one particular unit in the Israeli Army called 8200, which I should mention to viewers is constantly referenced by everyone in Silicon Valley with a lot of awe. But you've got to realize, 8200 is also one of the largest units in the Army.

Huh.

So it's kind of like saying you went to the, you know, to the NSA. There's a lot of people there.

Right, right.

But it's also a melting pot of great people. And you finish your service, and a lot of these people have, again, suffered a lot together, gone through a shared experience together, and so there's good, tight bonds, and that yields great companies. What I find quite interesting, though, about the Israeli culture is all of the startups Israel produces are almost all entirely execution-risk companies.

Mm-hmm.

So, it's basically, give us a hard problem, an obvious market, and we'll build it. So, cybersecurity's a great big deal in Israel. A company like Check Point, or Waze, maps routing, really hard problem. We're going to do a better version of that. Or Mobileye, you know, we're going to figure out, you know, we're going to machine learn basically models for autonomous vehicles. Hard, meaty scientific problems. There are very few market risk companies that come out of Israel. And by that, I mean things like Snapchat or Facebook or even Gusto or Airtable. You know, this latest breed of unicorns from Silicon Valley. And that is, I think, a very optimistic point about what California has going to it, because I think growing up there, the reason Israel wasn't creating those weird, creative companies and was more very focused on if you build it, they will come type stuff, it's because the culture is a little too realistic and a little too harsh. And to our earlier conversation, everything new and big starts off as small and stupid.

Right.

A lot of people told you know, the Airtable founder Howie, basically building an Excel competitor, that it was stupid and Excel had it all covered. A lot of people told the Airbnb founders that it was stupid, hotels were good enough. And in Israel, that's beaten down so much into you because the culture is so harsh. There's a, you know, there's a lot of like, it's almost like if you watch Seinfeld, you know, the kind of, really?

Really?

You're gonna build that?

Yeah.

And that's gonna work? No way. And so, That is the Achilles heel of Israel, is that I don't think it'll produce a market-risk company where it takes a lot of creativity, flair, and kind of the Californian optimistic view of the world to produce something like that. The reason this question matters so much, it— I mean, both as an Israeli citizen, I care a lot about it, but more importantly, the question is whether that same dynamic and contrast applies to China.

Right.

Hmm. Whether China is gonna be a country that's brilliant at executing. Give us a hard problem, we'll optimize it.

Throw tons of money at it.

Throw tons of money at it, tons of people at it.

Yeah.

But whether they're gonna be able to develop the weird, creative ideas that so far have been the lock on Silicon Valley.

So as a foreigner, what do you see about America that, that makes it exceptional in this way? One of my best friends in New York, he's from Ireland, and he— what one of the things that he loves about America is he says, You can walk down the street, at least about New York, you can walk down the street, sing at the top of your lungs, doesn't matter. And he dresses like crazy. Some days he'll dress like a girl, some days he'll just wear crazy stuff. And it's funny because he is like such a normal, like very creative, but when you're with him, very normal. But in terms of his dress and in terms of how he acts sometimes, just totally crazy. And he grew up in Ireland, played Gaelic football in Ireland. at the top level, and he said that America has given him this freedom to just let the floodgates open in terms of figuring out who he is and expressing himself. And that stuff, you know, you see it in the mundane in terms of walking down the street and what you wear, but I think that it 100% applies to startups.

Yeah, no, I agree. And I think America's greatest gift is basically there's a high degree, I think across the entire country, and I understand some areas are very challenged these days, but But in general, and compared to other countries, there's a high degree of optimism about the future. Again, I think California has this to an extreme degree, you know, and maybe slightly reduced in the Midwest. You know, maybe New York has it as well. But by and large, I think Americans are optimistic about the future. And this is important because they're also— that creates high degrees of trust and bonding in between people. And when you have that, you have efficiencies. That optimism can also cause 2008, because there's a propensity to believe that things will move actually faster than they do, and, you know, a little bit too much coffee and a little bit too much hype, and suddenly you have a massive financial crisis.

Mm-hmm.

And so, you can kind of view America as this manic-depressive child that goes through these massive phases of mania, massive phases of depression, and hopefully, you know, what we're learning over the course of the past 100 years or so with things like the central Bank and the Fed, interest rate control. We're learning how to give ourselves antidepressants so that our depressed phases aren't as bad, and hopefully that doesn't mean our manic phases aren't as great.

Yeah.

But when you— the amazing thing about the United States is that people generally trust you. People generally assume that you're gonna do what you say you're gonna do, which just creates higher efficiencies in the system, and people, in particular in Silicon Valley, are really optimistic about the future. I would say that that's Silicon Valley's largest export, is just optimism about the future. And you could really get funded here to do pretty much anything as long as it's moral. And it really doesn't matter who you are. And it doesn't matter what age you are. You know, I don't think it matters what gender you are now, thankfully, in part to a lot of the female VC funds. Hopefully it shouldn't matter what skin color you are, as long as you have a great idea, you know, places like YC will fund you. Yeah, and I think that's really rare. So I think that's the largest thing we trade in, is kind of the American enthusiasm.

On that note of American enthusiasm, I have 2 more questions for you, and the first one is John Rockefeller. What about him stands out as exceptional and motivates you and inspires you?

I, in general, love reading about It's about all the 1910 to 1940s titans.

So Carnegie, Rockefeller.

Carnegie, Rockefeller, Vanderbilt slightly earlier, Getty.

J.P. Morgan.

Yeah, all these guys. And the— what's interesting about it to me is I've been wondering increasingly if everything in humanity is like a giant sine wave moving in cycles, up and down, up and down, up and down. And the problem is that people only live 62, 80 years. And so we never see the whole cycle, basically, right? But as human longevity goes up, I think we will start to see the whole cycle, and people will start to say, oh, this is exactly like what was going on when I was a kid or when I was a teenager. So for example, today there's a lot of hatred— or hatred, there's a desire to inflame the tech community. The media is very anti-tech right now. Turns out, I mean, if you go read the papers from, you know, I was reading a New York Times the other day from, I think, 1850-something. This is the same thing about big oil. It's, boy, big oil is destroying the world. They're terrible. They're awful. They're evil people. And what you realize is these things go in hypes, because if you scroll the time machine 20 years further back, It's all enthusiasm about big oil. And so I think it's really important to read about the previous cycle, because again, we discussed earlier the failures of what happens when you predict too much by using analogy. But it is helpful to inform your model, because it is true that history doesn't repeat, but it rhymes. Something similar is going to happen. And so, you know, Rockefeller is interesting again, because he's Jeff Bezos of the time. Or maybe a slightly better one would be Zuck in the sense that— or Mark Zuckerberg— in the sense that when you read about him, he comes across as very diligent, critical thinker, brilliant strategic moves, not the most like creative person, but very brilliant reactionary strategist. Oh, you guys did that particular deal with that railroad? I'm gonna buy this other railroad and play it really well.

He basically played What was that story about the dirty oil, and then he hired the other person to go clean up that oil?

Oh, there's a general challenge that they had with— there was a— I forget the name of it. There's a particular oil that produced a terrible odor, and he hired a scientist, and they managed to refine it. And I don't think they got rid of the odor, but they found a market for it, basically, right? And it was, again, this is a great example of something that's treated as useless or a toy that becomes valuable. Like another— the most famous example of this that Clayton Christensen popularized is the examples of the mini mills taking over the rebar market and slowly eating away at the large steel mills. So basically, many of these mini mills were popping up. This is 1940s or so, '50s, and they were basically able to produce steel at a much lower grade than US Steel could, than a large mill could. And for a while, US Steel is ignoring them because they're like, those guys cannot produce high-quality steel. Mini mills find that they can produce rebar just fine, and it turns out in the rebar market, you don't care that much about the quality of the steel. And the rebar guys for a long time had been telling US Steel, look, we don't need your fancy steel, we just need simple rebar steel.

Right.

Please just set— and meanwhile, mini mills call them up, say, look, we'll slash the price, the steel's not as good, but you don't care. And they're like, sounds great. So they find a market. And over the course of 20, 30 years, they eat their way upmarket, starting with the rebar folks, and then slowly they get more cash flow from the rebar folks, and then they start serving larger customers and larger. And every single time, US Steel is like, oh, thank God we don't have to talk to those customers. They were too small for us. But small becomes big by volume. And so that's another example, just with, with that useless smelly oil, that great things start really small. And people really need to realize, because you look at Elon Musk today and you're like, wow, that guy is building stuff that is going to space. He is also building electric cars. Seems really big, really ambitious. Every one of those adventures started out as really small, stupid ideas. I mean, SpaceX started out as the Green Mars Project, where he was going to call up a couple of Russian friends, launch a bunch of rockets to Mars, take a photo of something green being planted on Mars, and then kill the company. And so the idea was the company will die, but it'll increase NASA's budget. Fast forward 10 years, the man is building the largest private space company, I think, in the world, let alone the United States. United States, does not start that way. Facebook starts out as a stupid directory in Harvard. Google starts out as a Stanford project. So you've got to realize these big things look really small at the beginning, and it's important to realize that, because once you do, you realize, I can work on a big thing, you know. Almost nothing starts out from the outset where the founder is like, I'm gonna build the largest company in the world.

Right.

People really get tricked into building large things, it's, it's that they slip into a feedback loop. They slip into a video game where it's like, you know, I started out building this small thing, people really loved it, started doubling down on it, and like lo and behold, you know, we started selling books on the stupid website called Amazon, but now it's working really well, so we should sell VHS tapes. While we're selling that, probably should sell DVDs. Fast forward, it's one of the largest companies on the planet.

Right.

It does not start that way. So I think it's really important to remember that, because when you're trying to figure out, oh, I want to learn Python because I want to build this The thing— don't hold yourself back because it feels small. That's how everything starts, right?

One last question. Yeah, and I'm really curious to get almost the most distilled version of how you think about making choices in terms of how to spend your time, what to pursue, specifically thinking along the vectors of productivity, personal growth, career growth, and fitness.

Well, I view fitness as the ticket of admission into the world, in the sense that I think it will just enable me to do more of all the other things I want to do. And so—

Run a couple marathons?

Yeah, I used to run a lot. Well, that's funny. What's funny is what I just told you is a little bit of a lie, because that's what I'd like to tell myself, but I have gone through phases much like you. I have a tendency to want to improve in every video game I engage with. And so I got into fitness with the general mindset of, I just want to be a better me and I want to help other people more. So I'm just gonna improve, you know, the level of energy that I have, my metabolism, you know, my overall strength. And then you, you know, you start running a little bit, you start running a lot, you start running one marathon, you told your friends you'd run another marathon, now you're running 4 marathons a year. Thankfully, I've dialed it back. I'll maybe run one marathon a year. Mostly because my body told me to stop, but I think that was unhealthy, actually, and I would not recommend that to anyone. I would recommend to everyone to treat fitness as something you do so that you can really play in the real world. For me, it's given me a whole other level of myself that I can access, and that's also made me very passionate about finding the next level after that. So I think fitness is kind of the base substrate you have to do. What that means, you don't have to go crazy and run marathons. It means that you have to like do cardio maybe 3 times a week, and you should probably be able to pull yourself up a couple of times. Invest in yourself so that you can invest in others effectively. So that's how I view fitness. Everything else, the struggle that I have is on the one hand, I have a desire to build a lot of these systems and games that I can optimize for productivity. How much did I read this week? How could I read more? What are my blood glucose levels? How can I improve everything? And then the challenge I have is that when I meet people that are truly great at the top of their field, all of these things that I am trying to do and apply active effort to, they're not really thinking about. And it's either that it's all in kind of system 1 in their head, it's just like the way they live life, or it's that I'm focusing on the wrong things. And so I think the main— one of the main things I'm very focused on These— you asked me a very large question, so I'll answer a simplified version of it. The thing I'm currently focused on is how can I get— how can I build the largest collection of good habits? And the definition of a habit to me is something you can do that does not require willpower.

Mm-hmm.

It just happens. How can I build a large collection of habits that are healthy, that are correct, and kind of save them to ROM in my head such that I don't have to think about them? And I really view that as— I'm in my 20s now, and I think it's harder and harder to change your brain and your body as you grow later on in life. And I would like to have that done, you know, by the end of my 20s. Like, I'd like to be in a good place in terms of body composition, in terms of what I eat, in terms of how I work, so that I can spend the rest of my life kind of not really thinking about all of that stuff and just kind of have it dialed in and figure it out. So that's, that's my current philosophy. And I do think that I think the thing people should do more so than really anything I write about in my blog posts is adopt this habit of self-experimentation early on in life. And it's important to do it early on in life for 2 reasons. One is when you find the right path, it's— you're more malleable, so it's easier to lock it in. And second is, as you mature in life, you develop responsibilities to others other than yourself, whether it's to your parents if they age, to your spouse, to your kids. And you really want that to all be dialed in. Like, you want to know by the time you're in your mid-30s, maybe you have 2 kids, you want to know, like, look, this is, this is like the workout stuff that works for me, and I can kind of just do it because I don't have time to optimize that and think about that anymore. Or if you're optimizing diet, you know, if you make a mistake, if I make a mistake with my diet, I'm one of those people that that could cost me 48 hours of productivity, of peak productivity. And I don't want to do that when I have kids. I'd rather figure that out now. So, I think the main strategy is kind of experiment early on in life, treat yourself a little bit like a startup, and then, you know, hopefully later on in life you kind of have— that investment will kind of yield dividends, you know, for the next 50, 60 years.

Awesome. Well, Daniel Gross, thank you so much for coming on the North Star Podcast.

Thank you for having me again. Cool.

Hey again, it's David here one more time. If you love to learn like I do, you can sign up for my weekly newsletter, the Monday Musings, share the coolest things I learn every single week, along with links to new articles and podcasts. You'll find a link in the show notes. At my company, North Star Media, we help companies and individuals build their brands. If you or anybody you know wants to start sharing their ideas, growing their audience, or building a brand for themselves, reach out to me directly. North Star Media specializes in blog posts, videos, and podcasts. Just like this one. You can support the North Star Podcast by leaving a review on iTunes, which helps people find the show. And if you enjoyed this episode, you'll also love the episode with Josh Wolfe, one of my favorite ones. Josh is the managing director at Lux Capital, which is another venture capital firm based in New York City. Now, in that episode, it's like an eclectic party of ideas. We discuss biology, randomness, Finance, consciousness, space, Brooklyn, investing, art, risk, and basketball. Man, I absolutely love that episode. Thanks so much for listening.