The Exchanges

Q I mean, I absolutely love it as a story. I mean, you mentioned PG's essays there, and I mentioned to you beforehand, speaking to PG before this interview, and PG did have one question. He said, how did you come to sell the company on eBay, and how did you end up making a decent amount of money from it?

A So we raised a very small amount of angel money, maybe 60,000 dollars in total, I believe, and Google Calendar had been released. And basically, because we didn't really believe in the startup as a mission, we didn't really have some deep insight nobody else had, Google calendar was pretty much the same as the product that we had built. And one of the results of that was that we kind of got demoralized and we weren't sure what to do next. And so we were trying to figure out how we might exit the online calendar business because we didn't really want to do that anymore. And we were tossing around ideas for how to sell it. And we, we knew we couldn't get like a investment banker or someone to take our, our little startup and represent it or sell it. And we knew we needed to get like interest and get people interested in buying. And so we thought it would be a clever stunt we hadn't seen anyone else do to list it on eBay. And we could get news coverage for it, basically. And if we got news coverage, it would act as if we'd hired someone to go sell it. And so we listed it on eBay, and then we went and got TechCrunch to cover. We gave TechCrunch an exclusive, I think, on our eBay startup listing. And we had set a reserve price of 50,000 dollars, and someone had actually bid. And we were so excited because we didn't think anyone would, like, Someone's going to want to buy the thing w…

AI assessment note: “we thought it would be a clever stunt we hadn't seen anyone else do to list it on eBay”

Q Okay, so my new year's resolution was to read more. So what's your favorite book and why should we be reading it?

A Favorite book. That's very hard. I read a lot. I love a lot of books. Right now, I would say Order Without Design. It's a book that I thought I was going to have to write myself, and then Alan Bertrand published it, and I was like, oh, this is amazing. Someone else has beat me to the punch on this thing. I get all this time saved, and it's about the tension between transit and housing and how cities work, and I found it very compelling, and I was like, oh, this explains this whole aspect of how it is that cities work, and I think that if you care about How to improve the modern American city or the modern global city, um, and you care about, you know, how your city works, it's worth reading Order Without Design, because it's, uh, just, I think it's really clear, detailed, data-rich explanation by someone who's an expert in the field.

AI assessment note: “Right now, I would say Order Without Design.”

Q Not at all, but I would love to kick off with a little bit about you, so talk to me. How did you make your way into the world of startups, and how did you become CEO at Twitch today?

A It's a long story. So I was in college with my friends, and friends with Justin at the time and my friend Matt And Matt pointed out that we were never going to have higher risk tolerance, and we were never going to have more access to other smart people than we did right men. And so if we're going to start a startup, maybe that was the best time to do it. And we found that compelling and started working on a calendar because we were all users of Gmail at the time, which had just come out. And we thought Gmail should have a calendar just like Gmail. And so we started working on a JavaScript powered calendar. And then we heard about YC, and we applied for that. And Matt wound up, as he wasn't an American citizen and wanted someone to be a consistent sponsor for his visa, wound up going to go work on Wall Street instead. But me and Justin carried on and started Kiko Calendar, which was just Google Calendar, basically, but before Google Calendar existed. And I think both Google and we looked at the same things for inspiration. We both looked at Outlook, and we said, man, Outlook's a pretty good calendar, but it'd be much better if it was on the internet and collaborative. And so we built a pretty standard calendar, but online. And in retrospect, this is a classic first startup startup, because it's not like, it's not like the calendar was something we passionately wanted to use as …

AI assessment note: “So I was in college with my friends, and friends with Justin”

Q I think it probably is a cycle. I do have to dive in and ask, because speaking of kind of resource allocation there, the thing that I always say and think is that the best CEOs are simply the best internal resource allocators. Would you agree with that suggestion?

A That is increasingly true as the company gets larger. So in the beginning, that's, I think that's absolutely not true. Like, I think that you're a six person startup. The best CEO is either an amazing salesperson or an amazing product leader. Or a brilliant engineer, and they, whatever it is that the company really needs, and they're the best at that company's most needed function, and they're really leading from the front and making it happen. And then there's some, there's some jump between, between the six-person startup that has to just get something out there to the 500,000 person megacorp that, where, where, where your job actually does turn into resource allocation, where you move from being a, I think it was sort of as from being a, a frontline general to a rear line general. Your job becomes about empowering other people to innovate and empowering other people to grow. And I think for me at Twitch, that happened somewhere around 400 people. It stopped, it stopped being something where my main job was to figure out what we should do and where we should go and how to meet the customer needs and started being more about deciding where to allocate resources and how. And that's where, how your resource pitch, right, how the pitch deck works and your method for allocating resources becomes absolutely critical because that's when your company turns into at some level and it b…

AI assessment note: “That is increasingly true as the company gets larger.”

Q I mean, I absolutely love it as a story. I mean, you mentioned PG's essays there, and I mentioned to you beforehand, speaking to PG before this interview, and PG did have one question. He said, how did you come to sell the company on eBay, and how did you end up making a decent amount of money from it?

A So we raised a very small amount of angel money, maybe 60,000 dollars in total, I believe, and Google Calendar had been released. And basically, because we didn't really believe in the startup as a mission, we didn't really have some deep insight nobody else had, Google calendar was pretty much the same as the product that we had built. And one of the results of that was that we kind of got demoralized and we weren't sure what to do next. And so we were trying to figure out how we might exit the online calendar business because we didn't really want to do that anymore. And we were tossing around ideas for how to sell it. And we, we knew we couldn't get like a investment banker or someone to take our, our little startup and represent it or sell it. And we knew we needed to get like interest and get people interested in buying. And so we thought it would be a clever stunt we hadn't seen anyone else do to list it on eBay. And we could get news coverage for it, basically. And if we got news coverage, it would act as if we'd hired someone to go sell it. And so we listed it on eBay, and then we went and got TechCrunch to cover. We gave TechCrunch an exclusive, I think, on our eBay startup listing. And we had set a reserve price of 50,000 dollars, and someone had actually bid. And we were so excited because we didn't think anyone would, like, Someone's going to want to buy the thing w…

AI assessment note: “we thought it would be a clever stunt we hadn't seen anyone else do”

Q Okay, so my new year's resolution was to read more. So what's your favorite book and why should we be reading it?

A Favorite book. That's very hard. I read a lot. I love a lot of books. Right now, I would say Order Without Design. It's a book that I thought I was going to have to write myself, and then Alan Bertrand published it, and I was like, oh, this is amazing. Someone else has beat me to the punch on this thing. I get all this time saved, and it's about the tension between transit and housing and how cities work, and I found it very compelling, and I was like, oh, this explains this whole aspect of how it is that cities work, and I think that if you care about How to improve the modern American city or the modern global city, um, and you care about, you know, how your city works, it's worth reading Order Without Design, because it's, uh, just, I think it's really clear, detailed, data-rich explanation by someone who's an expert in the field.

AI assessment note: “Right now, I would say Order Without Design.”

Q Not at all, but I would love to kick off with a little bit about you, so talk to me. How did you make your way into the world of startups, and how did you become CEO at Twitch today?

A It's a long story. So I was in college with my friends, and friends with Justin at the time and my friend Matt And Matt pointed out that we were never going to have higher risk tolerance, and we were never going to have more access to other smart people than we did right men. And so if we're going to start a startup, maybe that was the best time to do it. And we found that compelling and started working on a calendar because we were all users of Gmail at the time, which had just come out. And we thought Gmail should have a calendar just like Gmail. And so we started working on a JavaScript powered calendar. And then we heard about YC, and we applied for that. And Matt wound up, as he wasn't an American citizen and wanted someone to be a consistent sponsor for his visa, wound up going to go work on Wall Street instead. But me and Justin carried on and started Kiko Calendar, which was just Google Calendar, basically, but before Google Calendar existed. And I think both Google and we looked at the same things for inspiration. We both looked at Outlook, and we said, man, Outlook's a pretty good calendar, but it'd be much better if it was on the internet and collaborative. And so we built a pretty standard calendar, but online. And in retrospect, this is a classic first startup startup, because it's not like, it's not like the calendar was something we passionately wanted to use as …

AI assessment note: “So I was in college with my friends, and friends with Justin at the time”

Q I think it probably is a cycle. I do have to dive in and ask, because speaking of kind of resource allocation there, the thing that I always say and think is that the best CEOs are simply the best internal resource allocators. Would you agree with that suggestion?

A That is increasingly true as the company gets larger. So in the beginning, that's, I think that's absolutely not true. Like, I think that you're a six person startup. The best CEO is either an amazing salesperson or an amazing product leader. Or a brilliant engineer, and they, whatever it is that the company really needs, and they're the best at that company's most needed function, and they're really leading from the front and making it happen. And then there's some, there's some jump between, between the six-person startup that has to just get something out there to the 500,000 person megacorp that, where, where, where your job actually does turn into resource allocation, where you move from being a, I think it was sort of as from being a, a frontline general to a rear line general. Your job becomes about empowering other people to innovate and empowering other people to grow. And I think for me at Twitch, that happened somewhere around 400 people. It stopped, it stopped being something where my main job was to figure out what we should do and where we should go and how to meet the customer needs and started being more about deciding where to allocate resources and how. And that's where, how your resource pitch, right, how the pitch deck works and your method for allocating resources becomes absolutely critical because that's when your company turns into at some level and it b…

AI assessment note: “That is increasingly true as the company gets larger.”

Q Okay, really unfair question. Which of those individual skills do you think you're like, a fantastic at? And then which are you like, hmm, Emmett, I really need to spend some time improving this one.

A So I think there's one key skill that leaders need is the ability to be believed when they tell you something that that's what they actually mean. That they don't have a hidden agenda, that the thing they're saying is, is in fact... The real reason they're doing it, and if they, if it wasn't, they would just tell you, I can't tell you why I'm doing that, and I think I'm pretty good at projecting that. I'm pretty good at helping people understand what's going on in my head, so that when I say, hey, we're doing this because here's the reason, I'm generally believed, and I think that's something I've learned is, uh, really important. It's still, I still work on, but it's something I think I've gotten fairly good at. One area that I think a lot of managers are, I've seen be really good at, that I'm, I'm working on myself, is the ability to Enter a room on a topic you are an expert on, that you've really, you've gone deep on, that someone else is presenting in, and come in with full curiosity about what they have thought of and what they have believed, and not jump to the assumption of, you know what the answer is already. And it can be very easy for me to jump to believing I know what the answer is. It can be very easy for me to hear someone pitching me on a product idea for Twitch, where I've probably thought about this before, right? Like, I've been working on this product for a …

AI assessment note: “I think I'm pretty good at projecting that. I'm pretty good at helping people”

Q I absolutely love that you don't build communities, they grow. God, so many tweetable moments from this. I'm loving this one, Emmett. Tell me, who was the first check-in to Twitch, and how did it come about?

A Twitch is, in fact, Justin.tv. Like, we renamed the corporation, but technically speaking, I've had the same job since 2006 when we started Justin.tv, and I was signed on as a founder of that. And so the very first money in to Justin TV was Paul Graham, but at post pivot, we'd raised series a from all stop Louie. And then we'd raise some money from Silicon Valley bank as well as part of that. But I guess the first VC round after the pivot was with Bessemer and we struggled to raise that round. We had to pitch a lot of VCs, but Bessemer is amazing, but like we probably pitched four EVCs in that round. It was a struggle. Nobody wanted to invest in Twitch at first. We had some growth numbers, but it just seemed too ridiculous. I think.

AI assessment note: “the very first money in to Justin TV was Paul Graham”

Q kind of the first steps are the most important, but I saw a tweet of yours, and you said about it Starting businesses and really those early days. And you said it's like deciding to take on the burden of Sisyphus. I thought it was a really interesting analogy. What did you mean by this? And how does that correlate to the startup journey and flow in the early days?

A When I compare something to like the story of Sisyphus, I do believe the quote about Sisyphus in his boulder that he rolls up the hill that like, you have to imagine that Sisyphus is pretty happy. He has a purpose in life. You know, he's doing something. He's got the boulder, he gets it all the way to the top, it rolls down again. That's one more notch in the boulder rolling. And I think that what's different on a startup actually than the Sisyphean boulder is in the beginning, it kind of feels like it because you're, you're, it is, it is like Sisyphus. You're, you're rolling this boulder up and it's like just agonizing the hard work. Every time you roll the boulder slightly farther, it's like a ton of extra work. And then it turns out like there's a sling in your way and the boulder rolls back down and you start over. But there's this point, if you keep pushing it, where eventually you crest the hill, you get the boulder all the way to the top of the hill and you've been pushing it, you've been pushing it. And if you've only ever had a startup that doesn't have product market fit, you've never crested the hill because there's this point where the boulder gets up to the top and it starts rolling on its own downhill. And the first sensation is, oh my God, this is amazing. It's like, it's going on its own. I'm not pushing it. Like it's just rolling and it's, this is so easy. It's…

AI assessment note: “in the beginning, it kind of feels like it... You're rolling this boulder up”

Q Totally, it does at that point. I guess the question for me is, you know, you speak about having to embrace that resource allocation mindset, and if you're maybe self-reflective on yourself as your leadership journey, the question is, how have you seen yourself change and evolve as a leader over the last few years, and what elements have you found super challenging?

A So I found being a frontline leader a lot easier. I love talking with customers. I love the cut and thrust of detailed product design. I am an engineer by background and I like building things directly, and so I found that very natural, and I think that was the easiest part of my leadership journey was sort of leading a small team to build something where I could be working with other people, but directly involved in making the design decisions. Even if I wasn't making all the decisions myself, I was probably discussing them with people, and my growth edge over the past, I don't know, what would it be? Probably like six years since that became not true, maybe seven years, has been To learn how to inspire other people to that kind of work and let go of things. I think that good CEOs are often kind of control freaks early on because there's no one else to do it and somebody needs to make sure everything is going well. And then as your company gets bigger, you, you have to, I, I, I, I had to let go anyways. I mean, I found that was true for engineering. I had to, I had to let go of caring or controlling. I mean, I care about the engineering architecture decisions we make still. But I had to like, let go of attachment to them. I had to let me, it had to be okay for someone else to make that decision. And I had to let go of the idea that I would design every product that I would, I …

AI assessment note: “learn how to inspire other people to that kind of work and let go of things”

Q And then the final one, what are the next five years hold for you and for Twitch? Paint that roadmap for me.

A So for Twitch, I think the next five years is all about growing the base of the kinds of content that you can live stream. And interact around and making that really easy. Like we've lit up karaoke, so you can do karaoke through Twitch Sings. We're doing sports with Thursday Night Football. We've got some really interesting things happening around people chatting and just doing fitness and doing painting. And we're trying to, I think, find ways to add on top of gaming a bunch of other cool things you can do with Twitch. And I really do think the next five years is going to be the story of adding all of those things and expanding the community and welcoming more people in. For me, I'm hoping the next five years contains a combination of continuing to grow as a manager, but figuring out how to really empower other people inside of the company to own the stuff we're doing right now and the new things we're doing so that I can be involved as much as possible in the making of stuff. Because I, I really love myself doing, I love doing things like this podcast. I love creative work. And I think my journey of the next five years hopefully involves a lot of Getting to do, to do mentorship and getting to do work internally where I help us actually get stuff built. I'm looking forward to.

AI assessment note: “So for Twitch, I think the next five years is all about growing the base”

Q mentioned your explanation of how to structure an orbit. 1500 is the best he's ever heard. Now we both know he's had some pretty incredible founders. So that is quite a statement. So how can I not dive into this? How do you think about when you have crested that hill and you're pushing and it's sprinting down the hill? How do you think about structuring an orbit? 1500 people.

A Yeah. Cause when you're pushing the boulder up the hill, you want to keep your organization as small as possible because you never know where the top of the hill is. You need to keep your burn rate low because you You don't know how long you might need to be just inching that boulder up the hill to get to the top. But then once you start chasing the boulder, you really need to be able to, to go fast. And at that point, it becomes very much a game of execution speed and keeping up with the product pulling you forward. And for that, you often do need to scale the organization up and you start hiring people. And I think I have sort of two underlying beliefs about organizational structure. The first one is as you're scaling an organization is you're adding a lot of people specifically. I have two beliefs about organization structure. The first one is there is a fundamental tension or trade-off between centralizing and decentralizing, and those are really the only two moves. And if you centralize a function, you centralize engineering, you centralize product, you centralize customer support, you get people who can really learn from each other, you get mastery, and that organization develops a sense of esprit de corps and quality and what it means to be an engineer at Twitch. And those groups Can really hold a really high quality bar for the work being done in that area because they …

AI assessment note: “there is a fundamental tension or trade-off between centralizing and decentralizing”

Q like, the big question for me is, like, often we, you know, eulogize these founders who have these grand visions of always changing the industry, reinventing the way someone does something. How do you think about the balance between that unwavering persistence to push the boulder up the hill versus actually realizing maybe the boulder's not meant to go up the hill, and maybe we should work on something else?

A That is the judgment challenge for being a startup founder. I think for actually for any creative act, but specifically I can speak directly about, you know, making companies and products. You have to know when is this a sunk cost and I've discovered this can't work and I should give up and try something new. And when is this just a matter of persistence and grit and I just keep going, I'll be successful. And I don't think there's a magic recipe for that. I think it's, you know, that's a, that is almost the essence of a judgment call. If that was easy, you know, it would be an easy game and everyone would play. And I just don't think that Often we really know. And I think that sometimes people give up just before something would have started working. And sometimes people clearly persisted things well past where they should have, and they should have given up and tried something else. I do generally believe you want to know the bedrock first principle things about the world that you have faith in. So a good example of this might be like Amazon believing that commerce is going to go online in the nineties. Amazon clearly had a belief shopping is going to move on to the internet. How, where, when, we don't know, but like, we're very sure that's going to happen. And so you can just take all these different cuts at trying to make that happen. And if any given thing you're trying doe…

AI assessment note: “That is the judgment challenge for being a startup founder.”

Q Okay, really unfair question. Which of those individual skills do you think you're like, a fantastic at? And then which are you like, hmm, Emmett, I really need to spend some time improving this one.

A So I think there's one key skill that leaders need is the ability to be believed when they tell you something that that's what they actually mean. That they don't have a hidden agenda, that the thing they're saying is, is in fact... The real reason they're doing it, and if they, if it wasn't, they would just tell you, I can't tell you why I'm doing that, and I think I'm pretty good at projecting that. I'm pretty good at helping people understand what's going on in my head, so that when I say, hey, we're doing this because here's the reason, I'm generally believed, and I think that's something I've learned is, uh, really important. It's still, I still work on, but it's something I think I've gotten fairly good at. One area that I think a lot of managers are, I've seen be really good at, that I'm, I'm working on myself, is the ability to Enter a room on a topic you are an expert on, that you've really, you've gone deep on, that someone else is presenting in, and come in with full curiosity about what they have thought of and what they have believed, and not jump to the assumption of, you know what the answer is already. And it can be very easy for me to jump to believing I know what the answer is. It can be very easy for me to hear someone pitching me on a product idea for Twitch, where I've probably thought about this before, right? Like, I've been working on this product for a …

AI assessment note: “I'm pretty good at projecting that... One area... I'm working on myself”

Q Can I ask, does it create instability within the team and the group switching between these structures, which are so very different every two years or so?

A I mean, I think anyone who's been through a bunch of reorgs would say, yeah, they can very much cause instability. I think there's a lot, I, we, I hired a COO, Sarah Clemens, a while ago, and she's taught me more than anyone about change management in an organization, and what I've discovered is when you do a reorg, there's a way to do it such that everyone experiences it as chaos, and there's a way to do it such that it seems like the logical thing to do, and everyone understands what's happening, and it's just almost like, of course, how can we do anything else? And somewhere in between, that's where most real reorgs happen. Somewhere between pure chaos and perfect organization. But there's ways to do it that really do make reorgs feel less chaotic. But I think more to the point, it may be so that reorganizing your company, reorganizing a team, and centralizing a function that used to be decentralized, or decentralizing a function that used to be centralized, will cause confusion and chaos. And also, you just don't have any choice. You have to do it. At some point, you, you need the autonomy of decentralization, or you need the mastery of centralization, and You really just have to go for it, and you go for it as best you can, but that's a necessary and unavoidable consequence. And I actually think, at some level, it's good for employees, too, because you, you're shaking up h…

AI assessment note: “yeah, they can very much cause instability”

Q Now, this is an incredibly hard one. Biggest single piece of advice for anyone looking to build strong communities, Twitches is incredible. What advice would you have for someone looking to build their own community?

A So, I would recommend reading this book, The Art of Community, would be the first The first fast recommendation. The particular object level advice would be that you don't build communities, you grow them. Your communities are more like a garden than they are a building. And like a garden, it's important that you, there's things you can do to make the garden more likely to be successful, more likely to grow, but it's about creating the conditions under which a community can grow itself, not about making it directly. And the other thing is, like a garden, your gardener, you know, if you're trying to grow a garden or a woods or something, The plants you start with in the beginning are going to seed the whole thing. And so you start with a bunch of oaks, you're going to run with an oak forest. You start with a bunch of roses, you're not with a rose garden. And so those very initial starting members of your community, the people who make it up at the very start, determine in this very surprising way everything about it when it's bigger, and it just, it just keeps amplifying it. You would expect that that culture would shift more as you added new people who vastly outnumber the original people, but actually the founder effect On, uh, community growth is very strong, and so the number one question for your community is who are you starting with?

AI assessment note: “particular object level advice would be that you don't build communities, you grow them.”

Q like, the big question for me is, like, often we, you know, eulogize these founders who have these grand visions of always changing the industry, reinventing the way someone does something. How do you think about the balance between that unwavering persistence to push the boulder up the hill versus actually realizing maybe the boulder's not meant to go up the hill, and maybe we should work on something else?

A That is the judgment challenge for being a startup founder. I think for actually for any creative act, but specifically I can speak directly about, you know, making companies and products. You have to know when is this a sunk cost and I've discovered this can't work and I should give up and try something new. And when is this just a matter of persistence and grit and I just keep going, I'll be successful. And I don't think there's a magic recipe for that. I think it's, you know, that's a, that is almost the essence of a judgment call. If that was easy, you know, it would be an easy game and everyone would play. And I just don't think that Often we really know. And I think that sometimes people give up just before something would have started working. And sometimes people clearly persisted things well past where they should have, and they should have given up and tried something else. I do generally believe you want to know the bedrock first principle things about the world that you have faith in. So a good example of this might be like Amazon believing that commerce is going to go online in the nineties. Amazon clearly had a belief shopping is going to move on to the internet. How, where, when, we don't know, but like, we're very sure that's going to happen. And so you can just take all these different cuts at trying to make that happen. And if any given thing you're trying doe…

AI assessment note: “You're just giving up on the specific tactic you were trying.”

Q Can I ask, does it create instability within the team and the group switching between these structures, which are so very different every two years or so?

A I mean, I think anyone who's been through a bunch of reorgs would say, yeah, they can very much cause instability. I think there's a lot, I, we, I hired a COO, Sarah Clemens, a while ago, and she's taught me more than anyone about change management in an organization, and what I've discovered is when you do a reorg, there's a way to do it such that everyone experiences it as chaos, and there's a way to do it such that it seems like the logical thing to do, and everyone understands what's happening, and it's just almost like, of course, how can we do anything else? And somewhere in between, that's where most real reorgs happen. Somewhere between pure chaos and perfect organization. But there's ways to do it that really do make reorgs feel less chaotic. But I think more to the point, it may be so that reorganizing your company, reorganizing a team, and centralizing a function that used to be decentralized, or decentralizing a function that used to be centralized, will cause confusion and chaos. And also, you just don't have any choice. You have to do it. At some point, you, you need the autonomy of decentralization, or you need the mastery of centralization, and You really just have to go for it, and you go for it as best you can, but that's a necessary and unavoidable consequence. And I actually think, at some level, it's good for employees, too, because you, you're shaking up h…

AI assessment note: “anyone who's been through a bunch of reorgs would say, yeah, they can very much cause instability”

Q Totally, it does at that point. I guess the question for me is, you know, you speak about having to embrace that resource allocation mindset, and if you're maybe self-reflective on yourself as your leadership journey, the question is, how have you seen yourself change and evolve as a leader over the last few years, and what elements have you found super challenging?

A So I found being a frontline leader a lot easier. I love talking with customers. I love the cut and thrust of detailed product design. I am an engineer by background and I like building things directly, and so I found that very natural, and I think that was the easiest part of my leadership journey was sort of leading a small team to build something where I could be working with other people, but directly involved in making the design decisions. Even if I wasn't making all the decisions myself, I was probably discussing them with people, and my growth edge over the past, I don't know, what would it be? Probably like six years since that became not true, maybe seven years, has been To learn how to inspire other people to that kind of work and let go of things. I think that good CEOs are often kind of control freaks early on because there's no one else to do it and somebody needs to make sure everything is going well. And then as your company gets bigger, you, you have to, I, I, I, I had to let go anyways. I mean, I found that was true for engineering. I had to, I had to let go of caring or controlling. I mean, I care about the engineering architecture decisions we make still. But I had to like, let go of attachment to them. I had to let me, it had to be okay for someone else to make that decision. And I had to let go of the idea that I would design every product that I would, I …

AI assessment note: “my growth edge over the past... has been To learn how to inspire other people”

Q Now, this is an incredibly hard one. Biggest single piece of advice for anyone looking to build strong communities, Twitches is incredible. What advice would you have for someone looking to build their own community?

A So, I would recommend reading this book, The Art of Community, would be the first The first fast recommendation. The particular object level advice would be that you don't build communities, you grow them. Your communities are more like a garden than they are a building. And like a garden, it's important that you, there's things you can do to make the garden more likely to be successful, more likely to grow, but it's about creating the conditions under which a community can grow itself, not about making it directly. And the other thing is, like a garden, your gardener, you know, if you're trying to grow a garden or a woods or something, The plants you start with in the beginning are going to seed the whole thing. And so you start with a bunch of oaks, you're going to run with an oak forest. You start with a bunch of roses, you're not with a rose garden. And so those very initial starting members of your community, the people who make it up at the very start, determine in this very surprising way everything about it when it's bigger, and it just, it just keeps amplifying it. You would expect that that culture would shift more as you added new people who vastly outnumber the original people, but actually the founder effect On, uh, community growth is very strong, and so the number one question for your community is who are you starting with?

AI assessment note: “The particular object level advice would be that you don't build communities, you grow them.”

Q kind of the first steps are the most important, but I saw a tweet of yours, and you said about it Starting businesses and really those early days. And you said it's like deciding to take on the burden of Sisyphus. I thought it was a really interesting analogy. What did you mean by this? And how does that correlate to the startup journey and flow in the early days?

A When I compare something to like the story of Sisyphus, I do believe the quote about Sisyphus in his boulder that he rolls up the hill that like, you have to imagine that Sisyphus is pretty happy. He has a purpose in life. You know, he's doing something. He's got the boulder, he gets it all the way to the top, it rolls down again. That's one more notch in the boulder rolling. And I think that what's different on a startup actually than the Sisyphean boulder is in the beginning, it kind of feels like it because you're, you're, it is, it is like Sisyphus. You're, you're rolling this boulder up and it's like just agonizing the hard work. Every time you roll the boulder slightly farther, it's like a ton of extra work. And then it turns out like there's a sling in your way and the boulder rolls back down and you start over. But there's this point, if you keep pushing it, where eventually you crest the hill, you get the boulder all the way to the top of the hill and you've been pushing it, you've been pushing it. And if you've only ever had a startup that doesn't have product market fit, you've never crested the hill because there's this point where the boulder gets up to the top and it starts rolling on its own downhill. And the first sensation is, oh my God, this is amazing. It's like, it's going on its own. I'm not pushing it. Like it's just rolling and it's, this is so easy. It's…

AI assessment note: “what's different on a startup actually than the Sisyphean boulder is in the beginning”

Q mentioned your explanation of how to structure an orbit. 1500 is the best he's ever heard. Now we both know he's had some pretty incredible founders. So that is quite a statement. So how can I not dive into this? How do you think about when you have crested that hill and you're pushing and it's sprinting down the hill? How do you think about structuring an orbit? 1500 people.

A Yeah. Cause when you're pushing the boulder up the hill, you want to keep your organization as small as possible because you never know where the top of the hill is. You need to keep your burn rate low because you You don't know how long you might need to be just inching that boulder up the hill to get to the top. But then once you start chasing the boulder, you really need to be able to, to go fast. And at that point, it becomes very much a game of execution speed and keeping up with the product pulling you forward. And for that, you often do need to scale the organization up and you start hiring people. And I think I have sort of two underlying beliefs about organizational structure. The first one is as you're scaling an organization is you're adding a lot of people specifically. I have two beliefs about organization structure. The first one is there is a fundamental tension or trade-off between centralizing and decentralizing, and those are really the only two moves. And if you centralize a function, you centralize engineering, you centralize product, you centralize customer support, you get people who can really learn from each other, you get mastery, and that organization develops a sense of esprit de corps and quality and what it means to be an engineer at Twitch. And those groups Can really hold a really high quality bar for the work being done in that area because they …

AI assessment note: “The first one is there is a fundamental tension or trade-off between centralizing and decentralizing”