Full Episode: Emmet Shear, Co-Founder of Twitch / JustinTV with YC Co-Founder Jessica Livingston

All right, thank you. This is gonna be fun.

I was a big fan of Paul's essays in college. I was reading his essays along with spending just vastly too much time on the internet in general. But I read all of his essays and, unrelatedly, me and Justin and Matt Fong had decided we should start a company. Just sort of generically, like if we're going to start a company, we should do it now. We have the least to lose, we have the most access to co-founders. Our logic was pretty sound. Now would be a good time to start a company as we graduate.

We were all seniors at Yale together. And we should have started looking around for a startup idea. And Gmail had just come out. I'd been working at Microsoft the previous summer on the Hotmail team, which was a hot mess. It wasn't their fault, it's just Microsoft had decided to fund all of Hotmail with a team of like six people for like 100 million users. There was no possible way that product could have gone anywhere because it was underfunded. Microsoft had decided, 'Well, we bought this email thing, it didn't really go anywhere, whatever,' and they just stopped funding it. I was their person working on the spam problem, the summer intern. That was my job, I was the spam guy. I was integrating a vendor, so it wasn't like they were making me write an anti-spam engine, but I'm integrating the anti-spam vendor. This tells you what state the team was in. It was really not in a good place. I still think fondly of my teammates, I just... it was hopeless. It was like one of those things, in retrospect, they had this huge lead. Anyway, so Gmail had come out that summer and I got an invite that summer. I tried to show everyone else on the team Gmail and be like, 'Guys, this is a big deal, this is coming,' and no one was interested. No one was interested in Gmail at all, it was amazing. But I was really interested in Gmail and so when I was back at school using Gmail, I was like, 'This needs a calendar.' Obviously they made the email client, but Google doesn't understand this, they need a calendar that goes with this. And of course it turns out Google had already built the calendar, they just hadn't launched it publicly yet. But we did not know this at the time. And so we started working on that, that was our idea. We're going to go build a Google Calendar competitor effectively, but Google Calendar didn't exist yet. And we're working on this and our friends know we're working on this. And so I get a message from who was it in the computer science program, I think it was Jim Terry, basically being like, 'Hey, you like Paul's essays, right? And you're starting a startup, did you see this announcement? He's doing a startup funding thing.' And I was like, 'Oh no, didn't see that, I totally missed it. Oh wow, we should apply for that.' And so we did. I remember trying to explain to Justin this is a good idea and he was like, 'Oh this is great, they're gonna give us money, I'm sold.' Big fat $15,000.

That was enough to let us... Paul's theory was correct, that was enough to let us leave school on graduation, do this instead of getting a job. And Matt's from Australia so he dropped out and went to go work for I think Goldman instead because they could get him a guaranteed visa. I think it was a little too risky for him to... even Australians, the visa situation's hard.

And so it was just me and Justin at that point. And we came up and we interviewed at the Garden Street office. I remember we stayed in the La Quinta Inn and we got a taxi to go over to the Garden Street place, which is much nicer. We should have stayed closer to Garden Street. I don't know why we stayed, we were just trying to save money because we were broke college kids. Which in retrospect, actually, that was the right call. I look back at a bunch of things where I'm like, now I would never make that decision, but that was the right call at the time.

Yeah, so Justin's mom had a friend, I think through church, who lived in Roxbury. And so she got us a place deep in Roxbury in this house with no AC, which in Boston is a big problem. I never understood the whole food desert thing, we were in a food desert. The closest groceries and stuff were quite far away from us. So you have to either walk a really long way or take a taxi or something. I think we had my car though, so we could drive for it. So I think we went to Costco a lot in the car. But yeah, in retrospect, I don't know why we thought that was a good idea. It was cheap and it had space.

We'd been sprinting towards a demo. I was a Paul disciple and at the time Paul's big thing was like, you need to build a... running code is the key, build a demo, get something going, start learning from there. And so that's what we did. And we had the drag-and-drop demo running. I think it was the first time I'd ever seen drag-and-drop work in a browser. I'd never done that before. And you could, one time for each appointment, you couldn't move it a second time, but once you could pick up any appointment and move it to another place in the calendar and it would work.

We had 40 minutes.

Oh yeah. So yeah, we were just learning. I mean, this was all new for us. So we thought you need 40 minutes. And compared to a normal investment process, we're going to make our entire decision on a 40-minute meeting and then we're just going to say yes or no. It sounded already absurdly shortened.

Well, this was one of our insights, was to make immediate decisions. So that was always the plan, we were going to let you know that night. And I remember it was pretty much a no-brainer for you. We're like, you know, great founders and they built something, great idea, right? Let's do it.

Yeah, yeah. Which to be fair, it wasn't a good idea. It would have been if Google hadn't launched Google Calendar. But I think Google was always going to launch Google Calendar. I just remember walking out of the meeting and being like, 'Holy... that was Robert Morris.' And Justin was like, 'Who?' And I was like, 'I think that was Robert Morris.' I was a CS kid. I knew RTM. I knew about the worm. I was like, we just got interviewed by... I don't even remember anyone else in the room or anything that happened because I was like, that was Robert Morris. Oh my god.

Justin didn't... that didn't register. Justin's like, 'Oh, this is Robert Morris,' and just, you know, he's a professor at MIT, just like, 'Yeah, yeah, cool, whatever, that guy.' And I was trying to... I remember afterwards just being totally starstruck and Justin being like, 'What are you talking about?' Because if you're a literal CS major, you're not going to learn about it. You have to be really into internet programmer hacker history to know why he's so impressive. But like, yeah, as a result, I actually barely remember... I can't... all I think most of my memories of the rest of it are kind of reconstructed. Like I know you were there, Jessica, but I can't remember you being there. I think Trevor was probably there, I don't think... I don't remember him. I barely remember Paul.

Because I was just like, 'Oh my god, RTM, RTM, this... I can't believe he interviewed us.' So funny, he didn't even ask any real questions. He's not super talkative in interviews.

Wow, oh my gosh. Like I would have the same kind of flash of a person, but that's it.

Oh yeah. I mean, we were so excited. We didn't have a backup plan. Our plan was get YC. That was the whole plan. If we hadn't gotten it, I don't know what we would have done. I mean, to be honest, Justin had a job that he hadn't... he took the signing bonus from, and after we got YC he returned the signing bonus. Because like, I think in retrospect... I did not have a backup plan. And I don't think Justin was actually going to take his energy consulting job in DC. I don't think Justin would have actually liked that. I think that was sort of like to appease his mother or something, like, 'Don't worry, I have a real plan.' But we definitely did not have a, 'Oh, if we don't raise money from YC, we'll go start this startup some other way.' I don't think it's an exaggeration to say we would not have started Kiko without YC. We would have tried a little bit, but I think we would have given up quite quickly.

I mean, I mostly remember sitting in our very, very hot, like sweltering Roxbury apartment with no AC, like in our underwear, programming, and just feeling so alone and so depressed. Because I'd just come out of being in the best social environment of my entire life with all of my college friends, and now it was me and Justin. I love Justin, but he's not an entire social circle. And it's just me and Justin in this house, we don't know anyone in the city. It feels like we're never going to be near our friends. And our friends are all in New York basically, like some of them are in San Francisco, but most of them are in New York at this point. And they're all having a great time. They're all working their first year at Goldman with all their other friends who graduated, or at consulting firms, or they went to law school or med school, and they're having an awesome time. And we're in this Roxbury... it was so depressing and so sad. I had a couple weeks of just like dark, 'My life is never going to be fun again.' Oh no. And the one bright spot every week was Tuesday dinners.

Tuesday night dinner was the only social contact we had of any kind. In retrospect, if we just lived at like Davis Square or something, we would have been fine. It was a total own goal. There's a bunch of college kids right there. Cambridge is full of our friends and cool things happening. But we just decided to live in Roxbury. I don't know, it was not very smart.

Yeah, like it was full on. We were doing nothing but the startup, which was sad and depressing, but also we were doing nothing but the startup. There was no competing anything. And I drove out across the country, my girlfriend wanted to... I got back together with my girlfriend, which in retrospect, you know, I feel like that was probably a bad idea. Not because we weren't... it's just practically, her moving to Boston into this environment where I was already depressed and didn't have a social network, and she also wasn't... in retrospect, if we wanted that to work, we probably should have waited until I at least moved to Davis Square. That would have been a better idea.

Yeah, yeah, totally. 100%.

It caused a little bit of tension. I think more just the insanity of like, why would you bring a girl... why would you generate this extra stuff that we have to deal with? But he got together with Christine while we were in the Roxbury apartment too. So I think we both got back together... they met in college, but they'd broken up and they got back together, I'm pretty sure, while we were still in the Roxbury apartment. And yeah, we ate a lot of Costco lasagna. I made a lot of chicken and spring mix salads. And then it was quiche. I would bake a whole quiche.

No, no, no. I would make a pie crust and fill it with like a dozen eggs. And then I would eat that for four days. Quiche is like the easiest thing to make actually, because if you can make a pie crust or you buy one, you just crack a dozen eggs into it and some cream and whatever else you want to put in, and then you put it in the oven. You cut it into a bunch of slices and you have infinite food. I was just looking for things that were time-efficient, time-efficient for a lot of protein.

Oh, I remember the whole summer. I remember Paul Buchheit coming in and giving us a talk on Gmail and just being like, 'Holy, that's a real engineer.' I didn't understand what engineering was. And I remember telling Steve that you couldn't access a... it's so embarrassing... I think something like you couldn't access a database without Java. And what I meant was you couldn't access a database without a server-side thing, but I didn't really understand how programming worked. I think I fooled you because I was a good programmer because I'd been in the Yale computer science curriculum, which is a very intense, abstract curriculum. But they don't teach you how to do anything. I didn't know how any technology actually worked. It's really a computer science as abstract theory program. So like, how does a database work? How do you connect a program that runs on a server to a database? I have this vague idea you can do that, but what would that look like? I'd never done it before because that's dirty practical work and Yale doesn't believe in practical work of any kind.

Turns out Yale's actually right, which is their attitude is, you'll figure it out. That part's easy. We're going to focus on the really hard stuff. And when you have to figure out how to install a database driver for your Python program, you'll... it's fine. If you got to this program, you'll figure it out. And they were right. It took me all of six weeks to figure out how a lot of that stuff worked. Although it took me probably two or three years to get good at it. That's fine. I'm glad they focused on the more rigorous components because that stuff I would not have learned over the next two or three years if I hadn't gotten it in school. Also from the summer, I remember prototype day and sprinting to have something. And then I remember demo day and we basically built a calendar. I think we'd even launched and we had some users. And I remember no one... I think we presented for like half an hour or something. How long were those presentations?

They were pretty long. They were like real presentations with slide decks and demos sometimes. For demo day, I remember we got interest from Steve Wolfram and we got to go pitch him. He did not wind up investing because he did not actually correctly... did not think we were going anywhere on that startup. I also remember, I think it was his like seven-year-old son was in the meeting.

And I thought it was really charming. It makes me want to take Lucas, my son, into meetings when he's old enough. To be honest, he asked like two or three questions because Stephen gave him a... was like, 'Hey, do you want to ask him some questions?' His questions were better than the average VC associate and probably better than half of all the partner questions I've gotten. They were really... he asked good questions that were on point, that made sense, that followed. Because he's obviously a smart kid and he's been watching his dad ask smart questions in the meetings for probably a year at that point, so he knows what to ask. He's pattern-matched it. And I was like, oh, that actually made me feel better about it. When he was just sitting there, I thought it was weird, but then when he got to ask questions, at first I was like, 'What is this seven-year-old going to ask me?' Maybe he was nine, I don't remember exactly. But just thinking, oh yeah, that's a totally sensible, reasonable thing to ask.

What summer? Really not during the summer. Yeah, during the summer we were in Roxbury and we didn't see anybody. I liked Steve and I liked Alexis. We were drawn to them as sort of the other... Dan and Aaron Schwartz, I think that set of us, and I guess the Me guys, Zach and Zack and Chris. We all kind of... I think Sam was off pitching all the time, like was just not... didn't hang out with us that much because he was working on his startup. For him that meant deals. And we were sort of the set of people who I think were both very serious about their startup but also wanted to hang out and talk and have some fun as well. And so that group, we all sort of hung out quite a bit. I remember being particularly drawn to Steve and Alexis, but it wasn't until after the batch ended when we were like, 'Oh, now what do we do?' And then we had to be proactive. If we didn't reach out to someone to try to hang out with them, we were never going to see anybody.

Yeah, yeah, yeah. Not during the batch, we didn't. But after the batch, we all worked... we would go to Garden Street and work because it was just so much more pleasant than where we lived. And I think you took pity on us because now YC gives you $500,000 or whatever, so you can afford to get an apartment that's decent to work out of. Then with like $14,000, you're really trying to stretch it. Like, okay, fine, we're out of our office. But we worked out of the office together, that was a big bonding thing. But also we just were a little more proactive about hanging out and meeting up once a week, like getting beers on a Friday or something. And we really connected later. Once we sold Kiko, like 15 months in, we sold Kiko on eBay. And after we sold Kiko, we didn't really have that much to do. And so we spent a summer basically distracting the Reddits. We would at like 3:00 p.m. go over to their house with Chinese food and beer and start playing video games. And that was great, that summer was amazing. And Steve got his revenge though, then he sold Reddit like six months or nine months after that. And so when we were starting Justin.tv and we moved to San Francisco, Steve moved into our building and every day at like five, Steve would show up with a six-pack of beer and we'd start playing video games. It's like he got his three months of doing that to us back when we were trying to start Justin.tv. So we were always a good influence on each other. But we were ultimately a very good influence on each other because I think the number one thing that's hard when you're running a startup is, man, it is hard to keep going. If doing the startup isn't fun, if you're not enjoying yourself, it's just so hard to keep the faith and keep moving ahead. The Reddits, Steve and Alexis, I think were a big part of why we had the momentum.

Yeah, I mean, I think those things were... looking back at it, I could have gotten the $14,000 from other places. Like if we scrabbled, we could have put together a tiny round of money from friends and family. But we couldn't have put together that batch without the batch. I don't think we would have... the batch mates and that was the actually valuable thing we got. That was much more so than the money.

Oh, I'm so mortified you just asked that. Because as Emmett was talking about how hot it was and they were in Roxbury, I honestly... I delivered them to almost every other group, but they're mostly in the Cambridge area.

I know, but I could have offered to give them... I would have bought them an air conditioner. It didn't even occur to us to ask. The thought that we could fix our living situation... no, this is just how it is. I think at some point we did buy ourselves our own Costco air conditioners because they're not that expensive. It was depressing and at the time I don't think I enjoyed it that much. But I wouldn't... I liked that we... I would recommend to most people, maybe not to go quite so extreme as like go live in Roxbury, but moving somewhere where you aren't right with your friends and you're not... startups are hard and it's good to have the outlet of like every Tuesday or every Friday or whatever. But it would be better if you were working like seven days a week in the beginning, like at least six for a while. I mean, maybe not if you're older and you have experience, but we didn't have experience. We had no idea what we were doing. All we had was relentless energy. And so the more we channeled that into the startup, the better off we were.

So, Google Calendar launched. At first we're like, 'They can't take us, we've got them.' And then we increasingly realized we don't got them. We actually have no ideas. We don't even know... fundamentally, we didn't actually use calendars and we didn't actually have any idea other than the calendar that goes with Gmail. And so when Google launched the calendar that went with Gmail, we just didn't know what to do. And we also... this is, I don't know if you know this, Jessica, we had real startup ADHD over those 15 months. So in addition to building Kiko, we built YouLookFamiliar, a social network for families. We built a sound app, which is basically SoundCloud, where you could upload songs and they had a little thing where it was playing, you leave notes on it. We built this thing we didn't actually launch, things we called Secret Project, a combination sort of like Reddit today but Reddit plus GitHub, but we gave up on that. We didn't even get to launch. Myspace was the big social network at the time, so we built a sort of a Klout for Myspace. We also didn't launch that one. We built a spider... we calculated a bunch of Klout scores and we lost enthusiasm and didn't launch it. And we built Kiko. And in retrospect, the reason we kept building these other things and these other ideas, all almost all of which by the way are at least multi-hundred-million-dollar companies if not billion-dollar companies, is because in our hearts we knew that this calendar thing... no one has built a billion-dollar calendar yet. The calendar thing really wasn't a good idea actually. And so we kept running away from it but then going back to it because like, no, we're supposed to work on the real thing we're doing. And we weren't ready to actually pivot the whole company. And YouLookFamiliar, by the way, it's like Ancestry.com basically. The whole idea was you could get your family into it. And what we'd immediately started getting is feedback from customers, if we'd understood how to do it, is like, 'I want to add my dead family members too. I want to not just build the family tree of the living people but everyone who I know.' Which is like, that's what Ancestry.com is. It's a social network that includes dead people. It's like Cola as a social network. And we basically were like, okay, we're actually going to give up on Kiko. Well, what do we do with it? We've spent all this time building it. Somebody should pay us for this, was sort of the idea. It's a calendar, it works, it's a JavaScript calendar that does work. Surely someone wants a working JavaScript calendar. Who wants that? We have no idea who would want that. And we sort of intuited the thing you need a banker for normally, which is to figure out who wants it, to go show it to a bunch of people and run an auction. But the entire price wouldn't pay a banker's fees to get started. So how do we get a bunch of people to see it? And we're like, well, we basically need marketing. And so, well, if we list it on eBay, that's funny. And I bet we can get TechCrunch to write a story about it. That was the reasoning. And we were right. We listed it on eBay and TechCrunch did write a story about it. And that caused a bunch of other people to write stories about it. And that meant that everyone who might conceivably want to buy a JavaScript calendar heard about the fact we were selling it. Because if you got a story in TechCrunch at that time and on TechMeme and the other aggregators, you would absolutely reach the buyers. And that's how we reached Tucows, which is based in Toronto, who ended up actually buying it. And you didn't have to pay a banking fee. We just paid eBay their like 2%.