Building an API-Powered Platform

Hi there - it's me, Amy. I was so happy for Jon and Thierry to join me on the podcast because uhmmmmm yeah, as a non-developer working at HackerNoon, I had a lot I had so many questions about building an API-powered platform LOL In my mind, building an API-powered platform meant that you had a limited target audience. Why would you target developers who need one specific coding function done? What I didn't realize is that is very common and it saves developers time because they don't have to create it themselves. And of course, I was happy to chat about diversity in the workplace (at around the 22-minute mark). I love what Jon said about how there are two schools of thought - hire people who are all similar so they all get along or hire a diverse group of people that may have conflict but have a wider set of varying opinions and backgrounds. I agree with him when he says that the latter comes with a lot of problems. But anyway, check out this epi of The HackerNoon Podcast and let me know what ya think! Until next time, See ya on the Internet There's Big CEO Energy on this episode of The HackerNoon Podcast. Jon Dahl, CEO of Mux, and Thierry Schellenbach, CEO of Stream, talk to Amy Tom about starting their API-powered platform 😱 Listen to The HackerNoon Podcast on Apple Podcasts , Spotify , or wherever you listen to your podcasts . On this episode of The HackerNoon Podcast, Amy, Jon, and Thierry chat about: Why would you build an API platform with a narrow focus? 🤷♂️ (02:32) What were Thierry and Jon's very first jobs? (05:50) Waaaitttttt, so when you sell a company and start a new company that builds off of your original idea, is that like... cool? 🥶 (10:27) How many team members were on the founding team of Mux with Jon? (16:29) And how many co-founders did Stream have? (18:04) How did YOU learn your Management Skills? (19:15)🤔 Are opposing opinions important? (22:20)👊 🐠 VOTE FOR JON & MUX AT https://startups.hackernoon.com/california/san-francisco-united-states 🐟 VOTE FOR THIERRY & STREAM AT https://startups.hackernoon.com/us/boulder 💎 Get approved for 10-20x higher credit limit with Brex - https://bit.ly/37doqiH Connect with Jon & Mux: You can get connected with Jon at mux.com You can vote for Mux at https://startups.hackernoon.com/california/san-francisco-united-states Connect with Thierry & Stream: You can get connected with Thierry at stream.io You can vote for Stream at https://startups.hackernoon.com/us/boulder Learn more about HackerNoon and Startups of The Year: For more information visit startups.hackernoon.com To read HackerNoon stories, check out hackernoon.com COMPUTER-GENERATED TRANSCRIPT: [00:00:00] Guys, I went to my first party post COVID and it was an outdoor gathering with about 30 people, an outdoor birthday party. And I'm not going to lie in the first 10 minutes. I absolutely froze. And I was like, I can't remember how to talk to anyone. And it was very interesting. So I'm excited to see like how COVID like life after COVID progresses and getting to know people and meeting new people in person. Amy: And gatherings and socialization. It's so exciting, but anyways, this is the hacker noon podcast. My name is Amy Tom, and today I am joined with the Aerie and Jon, welcome to the podcast, everybody. How are we doing today? [00:00:44] Nice to meet you. Thierry: [00:00:45] Yeah, doing great. Jon: Afternoon noon. [00:00:56] You are the CEO of mux and the area you are the CEO of stream. And I wanted to bring you on today because you both have built an API powered platform as your business. And I'm very curious about this. So to start though, I guess I have some questions about. Building an API powered platform in the sense of like, why? Amy: I guess, because as far as I know, and you can correct me if I'm wrong because I'm not a developer, but an API is something like, that's a very specific thing that a developer uses. Performance specific function, right? Like it's a set of code that helps them to perform a specific function. So starting a company that acts on a very specific niche function. I'm curious about why you chose to do that. Jon, do you maybe want to. [00:01:49] Yeah. I think there's a really specific reason why I started mux and got into video, but I think the general answer to API APIs is the way I think about it is a software is getting so much more and more complex today than it was. Jon: I'm not a video game person, but if you think about video games, like 20 years ago, like one person builds a video game from start to finish. They did every, maybe 30 years ago, they did everything. And now it's hundreds and thousands of people working on these really sophisticated things. So as software gets more complex the more efficient and that people can make software it has a huge value. So if you guys are a way of. Bringing in additional capabilities in the software without having to build it yourself which is just huge need. [00:02:27] But isn't it like such a specific niche function that you would run or do with an API. And so I guess, like, why would you go so niche is my question. Amy: [00:02:38] Maybe another way to put it is I think actually APIs are They're just the right way to deliver infrastructure today. And so most software is built. So I don't think it's a narrow, like we have an narrow-focused New York video, but the overall category of KPIs is just really big, important part of pretty much all sides. Jon: Yeah, I think Twilio used [00:02:57] to get that feedback as well, right back into their site Lakeway. They're doing this niche thing. It seems small. But yeah, that's the big trend in the market. So instead of building apps from scratch, people are using APIs from everything, from what we do here at stream for activity feeds and chat. Thierry: Two things like video search location, every component of an app is being turned into an API and then into a standalone business. And yeah, it enables customers to launch their app faster, to grow faster without the traditional scalability concerns. Some of the reasons that I started them. So my first startup was a social network. It grew to millions of users, and we had a lot of difficulties like keeping the activity feed up and running. To follow other users. That company ended up selling it did. Okay. And I just kept them thinking about it. Like, why did we spend so much time building something that tens, if not hundreds of thousands of apps have and that was a lot of the motivation for starting streams. So we offer these reusable components for building activity feeds and a, and chat. It's a really exciting time for engineers and for developers, because they're really good at some specific part of apps. It gives a really strong opportunity for people to start startups that maybe it wasn't there before. So I think it said the thing is pretty excited. [00:04:11] so essentially you're saying like giving them the development skills that they don't already have Amy: [00:04:16] Yeah. Yeah, it gives I think a lot of companies can build things in house, but we did that back in the day, but it's like time and effort and our mat, our messaging was like, you send a message, an email arrives. Thierry: There was like, no, there was no features or functionality in there whatsoever. And I think nowadays consumers just they expect. High level of Polish both in consumer apps. If you're using a dating app, you expect everything to work smoothly on the chat side of things. If you're like talking to your teacher, education is a very big vertical for us. So patients teachers, these type of communications, they should still work well. And that's what consumers have come to expect. I think that's a big part. It's driveway companies like stream and companies like monks are becoming very public because the bar just went up in terms of the quality that you have [00:05:04] to ship. Amy: Okay. Okay. Cool. Cool. Yes. Now I'm under getting to the understanding that like, this is more of an, a. Industry standard than I assumed it was. Okay, cool. We're on the same page. So now I want to take multiple steps of back and find out what your first jobs were, because I wanted to understand like how you got to where you are today. So the airy, what was your very first job? [00:05:27] I started the first thing I did was a gaming website when I was 13. Yeah. Wait, your Thierry: [00:05:33] first job was a gaming website. You weren't like a paper boy or something? Amy: [00:05:39] I think I did some, I worked in a supermarket and local supermarket for a bit as well. Slicing bread, all the good stuff, but yeah, Thierry: [00:05:46] like a more like quote unquote normal first job. Amy: [00:05:49] Yeah, and I think that I did that afterwards section though, the first thing I did was a gaming website and put outwards on it and started getting checks, which I'm from the Netherlands. So it was weird to get a check. I had no idea what to do with a check. So yeah, no, that was one of the first things. Thierry: Yeah. [00:06:04] Wow. Okay. And Jon, what was your very first job? Amy: [00:06:07] Very first. Like my paperboy job was. I had three or four lawns in the county. I grew up in that I moved once a week. And that was it. Jon: [00:06:16] How much were you making? Amy: [00:06:17] I think I got some of these were big lawns. I got 40 bucks, which was, I know. Jon: Yeah. Some were smaller but yeah, it was about that. [00:06:24] All right. Cool. Cool. And then from lawn mower, where did you go? Amy: [00:06:28] I kinda, sorry. I went to college. I actually studied philosophy in college and actually got a graduate degree in theology. So I proceed this totally different course when I was young. Jon: Okay. Wait, [00:06:39] wait, what did you intend to do with that degree? Amy: [00:06:41] Maybe something academic, maybe each right, that kind of thing. That was always something that attracted me. But then like I graduated and I'm like I should probably work for a few years. You don't get paid a lot of money as a philosopher or as a theologian. Jon: So I I actually, I taught myself programming and built a little dab shop. And that was my first like full-time job. Okay. [00:07:00] Can I ask around what year? This was Amy: [00:07:05] 2005. Jon: [00:07:06] Okay. So how did you learn programming? Amy: [00:07:09] I, I took I took a Paskal class in high school. And I took a little bit in college. Jon: I did like half of the CS degree, but really I just taught myself, like I, I found problems and I like read books. I worked through a bunch of books and a bunch of online sort of things. And I went from being a pretty bad programmer to a pretty medium programmer would [00:07:29] say. Okay. All right. Amy: And then how did you get on to start. [00:07:35] Yeah. 2007 I was running this little dev shop and we got hired by a startup that wanted to work with video. So they wanted to build like a better YouTube in 2007, which since YouTube is YouTube, obviously this do not work out. But I didn't know anything about video and they were like, okay, Jon, you could build a video translator. Jon: And I'm like, what's video transcoding. So I figured it, I figured this out. And the first thing I did was like, look for a vendor, look for an API, honestly, that would do what we wanted. And there was nothing out there in 2007. So I did it. And then, the startup didn't go anywhere. And then I I bootstrapped around this idea of video's probably going to grow. And this is a really hard problem. So I built a couple of rudimentary versions of a video transcoding cloud API service got together with some friends. And then we started a company called which was my last company, which was an early API bit early video API company. So that's the backstory. We ended up selling the company really young, like two years. Then we sold the company and at the end of that whole journey, it was like, man, that opportunity is still there. And that's why we started. [00:08:40] Okay. So wait, is Zen coder different from the service that mugs. Amy: [00:08:45] Yes, Zen cutter, Zen critter does one specific thing. Jon: It does video transcoding, which is something that everyone has to do. Like every video you ever look at on the, on, on the internet is transcripted. But it's really only one part of the picture. Mux does that plus a lot more. So max is it's one, one level higher of a service. And it does a lot of other things to you. [00:09:04] And I guess, like I have a logistics question about being a founder when you sell a company and then. A new company that's based off of the original company is that, I guess that's allowed cause you did it, but yeah. Amy: [00:09:18] Yeah. So I'm not a lawyer consult your lawyer, all that, but it depends by state. Jon: So we are in California, California has a very robust like pro startup pro employee. Thing non-competes are generally not enforceable in California except during an acquisition. So we actually did. We signed a non-compete and we honored a noncompete. After selling the last company those expire and and then you're more free to [00:09:43] do it. Amy: Yeah. Because as a founder, is the point of founding a company not to like eventually sell it for a lot of people and then just start another one. So I think that kind of makes sense. Okay, cool. Cool. Cool. And so in theory, I would love to learn more about what happened after the stint in the grocery store. Where did you study? [00:10:06] I studied business at Erasmus university. So that was fun. International business. We had people from all over the world. I started on my first startup when I was 20. So the last year in university, I was doing both both working and studying at the same time, which I don't recommend anyone to do. Thierry: It's a hassle. But yeah, the first company ended up ended up growing sold. Then for stream, we participated in Techstars New York because one of the things that really frustrated me with the first company, how hard it is to raise money in Europe. No. If you've had other founders talk about this, but raising money in Europe can be quite difficult. And it, we did Techstars in New York and I still remember at Dartmouth. From HubSpot, like straight after the program emailed me and he's Hey, I'd like to invest a hundred thousand. And I emailed them back okay, do you want to jump on a call or talk? And it's no, I'm busy, but sign me up. And basically like he ended up investing before the paperwork was even done. It was great, like raising money in the U S as much. I think the VC ecosystem here is very sophisticated. And that really is an advantage. So that's why I moved to Colorado. We ended up raising our first round second round third round here. The last two rounds we raised during the pandemic. So that was an interesting twist on things where you don't meet up with the feces that's but of course, growth accelerated for us quite a bit with older chat experiences, moving online from like life event companies like hop into education to healthcare. Everyone suddenly needed to have a Polish chat experience. So that that that helped at the moment we power activity feeds and chat for a billion end users. So the alts are pretty high to some of the opting or phone or. [00:11:35] Okay. Wait, I have some questions about venture capitalists things, and you're up for that and because how what's the main difference would you say? Amy: [00:11:45] I think the main difference is that you typically need series a metrics to raise a seed round and in Europe it's just one stage harder for every round. It's getting better. So there's some new funds coming and they're starting to be more, more aggressive, but I still see. Most companies in Europe struggled with the fundraising side of things or raise sometimes to succeed, to raise from us investors. Thierry: But it's difficult if you're not in the country. So yeah, it's a tricky problem. [00:12:07] Interesting. Okay. I did not know about that. Cool. And wha I guess, like, how did you stumble upon starting your company? Amy: [00:12:16] We just ran into this problem ourself. My first startup hat and activity feed, it had messed up. Thierry: They both struggled in different ways. The activity feed was impossible to scale. The messaging was not as polished as you're used to from like an iMessage or WhatsApp or a slack. So yeah, we were solving our own problem. And we saw the bigger trends would companies like SendGrid, they do it for email Algolia they do it for search. So this trend started becoming a thing. And I think in 2014, 2015 we got started with with stream. Jon was said max, from around the same time or are you guys been [00:12:46] yeah, 2016 really, but pretty close. Yeah. Jon: [00:12:51] So what I think is interesting is that Jon, you come from a like software background or, you taught yourself how to do the software and that's how you got into it. Amy: And the area you came from a business background. So I guess, do you feel like one has more weight than another in terms of being a founder and CEO? Jon [00:13:11] actually, I started learning how to code at 13 Thierry: [00:13:17] developer. Yes. Yeah. Amy: [00:13:19] I think everyone learns how to code most of us, at least two. Thierry: [00:13:23] Yeah. I think the conventional wisdom now, like 20 years ago, the conventional wisdom was you got to have an MBA. If you want to start something right or not, I was doing this. Then now the conventional wisdom is you have to be a technical. Jon: You have to have at least one technical founder and maybe all technical founders. In reality. I think the reason is not that you actually want your CEO of programming, but if you're building software, if the domain of software is just really valuable to understand the domain. So it's not there. There's lots of successful companies that don't have engineer founders, program founders. But I think these days, probably most of you [00:13:56] most do. Yeah. Yeah. At least if you yourself are not the technical co-founder, then you have another one. Jon. You are a co-founder though. And so are you theory, is that correct? Amy: So how many founders are part of mugs? Jon? [00:14:12] There's four of us which is the big founding team. Usually you see two or three, but yeah. Jon: [00:14:18] And everybody has a different role and part of the organization, Amy: [00:14:23] yeah. Yeah. Yeah, we're all technical. I'm, probably by far the worst engineer out of all the whole group. Jon: But yeah, w one, one actually has, is leading. We call it developer experience, but it's really marketing plus support plus solution engineering, dev advocacy. Open-source all that. Just cause yeah and then one leads product and one came from a led engineering for a long time [00:14:45] for us. And what was your. Amy: [00:14:49] See ya. So the big boss it's fundraising. And then that does where it starts, like your first round. It's okay, someone's got to do that. And then lots of things happen. Jon: [00:15:00] And how many people are a part of your team today? Amy: [00:15:03] I brought 110. Whoa. Okay. Yeah. And it's new to you. We were probably half this a year ago wow. Jon: [00:15:11] Okay. So growth is going crazy. It must be stressful to manage as a CEO. Amy: [00:15:17] It is. Although it's always stressful. It's like people, you feel the same burden as at least I felt the same burden at 10 is a hundred. It's it's. I'm I'm sure. Maybe it's different at a thousand, but Jon: [00:15:27] Okay. Amy: And the area, how many peop, how many founders did you start with? Yeah, [00:15:32] so it's a Tomasa and myself. I'm technical and S focused more on the marketing side of things, and Tomasa is like really deep into the storage technology and the tech side of things. So Tomasa is the more technical version on the team. Thierry: And that's been it's been a crazy, like the last a year ago. Streamless. We were about dirty people. And now we're up to 130, so it's been [00:15:54] indeed a Jon: [00:15:54] little stressful as you say. It's interesting. It feels like it changes when you go from managing Thierry: [00:15:59] people who do to work to like a couple of layers in between. Jon: [00:16:02] I dunno how you experienced that jump, but it's, I dunno for me, it's a different Thierry: [00:16:06] five. Yeah, for sure. Jon: [00:16:09] Yeah. Yeah. Okay. And I guess, like, how do you decide to hire that many people at once? Going from 30 to 130 and 50 to a hundred and less a year or less is very fast. So how do you decide that? Amy: That's the direction you're going to go into? [00:16:28] Yeah. Jon: [00:16:29] Sorry, do Amy: [00:16:30] you want to Jon: [00:16:30] start out? I was just about to say that I think it's very we were very conservative with our growth or like the first four or five years. But then because the metrics that they just been justified doubling down that aggressively but then our growth really accelerated as a separately barking feeds and chat for a billion engine servers, our chat Thierry: [00:16:49] product grew, I think it was like 625% Jon: [00:16:52] in one year. Thierry: So well, because it was like gross. We ended up raising funding and we also really needed to grow the [00:16:58] team because there's just even the customer support side of things. Like we would get maybe 50 questions Jon: [00:17:03] from customers in a month. Now we get like 700. So there's just so much Thierry: [00:17:06] more pressure on the Jon: [00:17:07] organization and you need a bigger team because of that. Thierry: But for us, we waited quite a bit before we saw the growth up before we took that step to increase the size of the team. [00:17:16] Yeah, I'd say not too different. We try to, it'd be disciplined with our growth. So every time we, basically, every time we raise around a funding after our second round we put together an operating model and I know this sounds old school, but I have an old school, like business advisor who really pushed me to do this. Jon: And it's the right thing where we really have like projections and try to have an understanding of what the financial picture is five years. And our goal has always been to try to hit positive cashflow on every round of funding. We're always on a path where, we can stand on our own feet. We might still raise more money and we did that a couple of times, and then we can just grow faster because now also we can hire a lot more people to get to, a profit but it's, and then on the other side, it's just like huge conviction that this is a big market. I, I. I have no doubt that the faster that we invest in the faster that we build, like a faster, that will for a long [00:18:00] time. Amy: If you are ready to scale your startup faster, you can get approved for a 10 to 20 times higher credit limit within minutes with VRX. So you can visit the link in the show notes of the episode, or visit brex.com for more information back to the episode. Okay. Exciting. So I guess I also want to know, what do you wish that you had known that, you know now when you're starting your own company, like if you had some advice to give to a future entrepreneur, what would it be? Jon? [00:18:31] So I'm gonna, I'm gonna give, I'm gonna give the opposite advice. 10 years ago, I would have given advice. That probably sounds I dunno Y Combinator advice or kinda like all the conventional wisdom today. Now, what kind of, what, which is really like. Build good software and talk to developers, which is totally true. Jon: Your conductor customers. I'll say the thing I've learned that I wish I knew was actually, as you scale a company and scale a team some of the like traditional stuff really matters. Like I used to think at my last company, I didn't know how to manage people. I thought management was stupid. It's I don't know, like what, why would you want managers? They're just like, dead weight or something or overhead or whatever. But actually a lot of just the fundamentals and like good management are just really important. And it doesn't come obviously to someone who's like a self-taught engineer, like it's a different track. But it's really helpful to bring the two tracks together. [00:19:15] Okay. Then how do you think that you learn to your management style? Because it's really hard. Yeah. It's really a skill. I'm only just starting to be calm, like even any kind of semblance of a manager. I now have a podcast editor shout out to Alex who helps me with all the podcasts. Amy: So I'm managing the podcast and I am very overwhelmed. Yeah, like how do you, how did you learn how to manage? [00:19:38] Yeah, I think I originally learned just intuitively and I got some things right. And got some things wrong. Yeah. Most of what I got right was around like good open, honest, transparent communication getting everyone like going in the same direction in a kind of team collegial sort of way that it wasn't. Jon: I was my intuition. I'd say over the last few years, I've actually learned by reading some good books and like trying to study some of the actual science behind. How do you build a successful team and a healthy team and a team that works well. [00:20:08] Okay. Book recommendations, Amy: [00:20:09] please. Yeah. Jon: So one, one is what I'd seen forever. But never read it. It's called five dysfunctions of a team. It's one of those books that like, you look on the business shelf and you're like, oh, it's probably some BS, like fluffy book or whatever. It's actually really good. It's by a guy who writes a lot on Teen health and company health. And I think it's actually really like a foundation of good management and good company culture. And then interestingly, the guy, I dunno, he wrote the book maybe in the two thousands or something. And then Google did a bunch of research or the last 10 years on what makes healthy teams and successful teams. They looked at like their highest performing teams. I think it's called like project Aristotle. Maybe you can Google it. I think and what they found that makes the most high performing teams was actually like, really exactly what this guy, who's a management consultant saw through his experience and observation. So it's really, I thought it was really interesting. It's actually like an empirical grounding, like a scientific grounding of. At least in this, at least in software, in this era, what makes teams healthy and successful. [00:21:06] Okay. That I would like to know your opinion then of what makes teams healthy and successful since you are a CEO. Amy: [00:21:13] Yeah. I'll just kinda run through the framework. And it's really does govern a lot of what I think about so say healthy teams start with a layer of trust. So if people don't trust each other, You're, nothing's going to work. On top of that layer of trust you need open, honest discussion, including people disagreeing, including debate, including healthy conflict. Jon: Like one of the things Google found is like their highest performing teams. Every person on the team spoke up and on lower team, they were like dominated by one person. So you need that foundation of trust that lets everyone speak up. It gets the best ideas on the table. People can disagree with each other and it can say that's wrong. And then from there it's like making clear decisions and saying who does what? So everyone's clear. Okay. I might disagree with it, but like we made the decision and like you're responsible for this thing. So deciding then being accountable and then making the last thing is just making sure that you're, everyone's aiming towards like real goals, like the same goal. If people are doing, if one team is making a decision or one person's big decision based on like their own ego or agenda, and someone else is trying to do it by what's best for the company like that doesn't work, you really need everyone focused on the results. [00:22:20] Okay. I don't know how to ask this without making it like a leading question, but do you think that it's important to have diverse hiring then because having opposing opinions is important. Amy: [00:22:33] I definitely do. It's actually two schools of thought on this. There's one school of thought. That's you want everything to be the same, so you have no conflict. And so you don't have to worry about like interpersonal relationships. I don't know, maybe that works, but I think you run into a lot of problems that way. Jon: I think if you don't want to do that, like everyone, you're only hiring your friends then. Absolutely. I think, and this is what we try to do at max. I think it's really valuable to it's both valuable to have a diversity of perspectives and that's all of the, the traditional types of diversity, as well as diversity of background and company you don't ever encounter from the same company, those kind of things. Yeah, I think it's, I think it's really important to get the right ideas on the table and actually the work that you do, the work that you do to be like an inclusive place. That's like welcoming a lot of people. I think it really overlaps with the work that you need to do to build a culture of trust and safety, which is that foundation. [00:23:20] Yeah. Yep. Yep. I agree. And when I say diversity, I really mean yeah. Diversity of background, of skills of education of race and color and gender and everything like that. Yeah, I think I personally think that it's important and I would rather, if I was going to be a CEO, I think I would rather hire people who are going to argue and disagree with each other, then everybody who's just gonna be like, that's great. Amy: [00:23:43] Yeah. Yeah. I think, that there's more than one way to succeed. Really toxic companies that succeed, but that, that, what you're describing is what is what I want to do. Jon: [00:23:53] Okay, cool. And so at the area, I'm going back to you now, what do you wish you knew before you started your company? Amy: [00:24:02] All right. Yeah. I love things, but I think one of the honest. Aesthetic business and for many years, so we built these books and I thought it was very hard to translate that into the day-to-day running running a startup, one book recommendation that I think was excellent and Thierry: [00:24:19] really helped us structure Jon: [00:24:21] How we run the business. Thierry: And it's this book called track. You should have a look at that. I think it's an excellent book. I liked how it distills a lot of the management best practices down to something simple that the whole team can understand. So that's an area that we've been very focused on, like making sure that we have very regular meet ups about the KPIs, but the progress on main goals. We recently implemented this tool called lattice to make sure that all will go, all teams have clear goals and that we that their managers have a framework that ensures that they do all the management best practices. Because one thing that we noticed is that so many of our team members. Being the leader or the individual contributor on the team to now having routed it and relatively large teams reporting to them. But they often don't have a [00:25:02] lot of experience managing larger jobs. Jon: [00:25:04] So I'm feeling right now. So Amy: [00:25:08] yeah, for us just came in came in quite helpful because it's things like one-on-ones performance reviews tracking your goals. Thierry: I also thought that. So one of [00:25:16] our investors Dharma shock, he wrote the speech. You can Google it about like vectors and aligning the Jon: [00:25:20] organization. I'm not sure that one's working towards the same goal. And I actually think that when you get past say 30, 40 people, that becomes one of the hardest things. Thierry: Like it's just, it's about aligning people to make sure that they're working towards the same goal and then recruiting. And yeah, it's funny because if you're getting started with with a new company, you're so focused on the product, you're focused on the engineering side of things. But then at some point you need, it becomes all about entrepreneurial as well. [00:25:43] Yeah. Okay. So what, at what point of a startup's life is, are you going to be putting in like management process? Amy: [00:25:52] A little bit at the beginning, but then a lot more after say 40, 50 people you get more structured around defining all the process all the processes in your company. Thierry: You get more structured around goals. You switch from Trello and Excel based like tracking of things through like proper [00:26:08] JIRA like everything gets more formalized Jon: [00:26:10] a little bit around like 40, 50 people, at least that's how it was how it was for us. Thierry: [00:26:14] Was that the same experience? Yeah. Amy: [00:26:16] Yeah, pretty similar. I think the there's a stage of a startup that's just trying to get to product market fit. And at that point you can be pretty processed. Like you can be pretty management light, you can pretty flat, like you're trying to get a product that customers love and will pay money for and make that repeatable. Jon: But then the next stage is like, how do you grow and scale that? And I think that is, for us, it was probably around 40 or 50 people too. That's the point at which you have to really invest in good management, good executive leadership, which is different, like management and executive leadership are two different skills, all those kinds of things. [00:26:49] Yeah. Okay. So I would love to talk about the scaling process then, because as a non CEO, I think this is the part that like, really apps with my mind, because it just seems like so complicated and so much like big picture. So the area, what are some of the things that keep you up at night about scaling your business? Amy: [00:27:08] A couple of things. I think the it's very hard to make sure that your eight guys are flexible enough to support all the different use cases. So we spent a ton of time trying to make sure that any possible chat or messaging or activity feed use guys actually fits our API model. Because the value proposition. Thierry: Selling an API is that it's for usable. So in theory, we can spend, a thousand times the time and resources on building something, then like our customers can just because we get to reuse the same bit over and over again, and then focus on optimizing it. But that only works if you get reusability rides and it only works if you get the reliability and performance. So those two things I spend a lot of time thinking about and yeah, those are the main. [00:27:52] Okay. What about you, Jon? Amy: [00:27:54] I'd say the things specific to API businesses is just that our, a mission critical part of your customer's business. And you have to be perfect, which is really far. Jon: And what one is perfect because software [00:28:06] is not perfect. Yes. If you F up your reputation. Yeah, Amy: [00:28:10] exactly. And your customers might have their entire businesses on the line. It might just be your employees might be their employees too. Jon: [00:28:17] Yeah, actually that's not something that I put too much thought into, but as a startup, I guess you have to have an amazing track record as well. Amy: Otherwise people will not believe you are invested in you, okay. Good building off of that. How did you get people to believe in you when you first started Jon? [00:28:37] The first time was Zen coder. It was it was just building a product that that customers really liked. It was good design co both, partly visual design, but also just like product design, documentation, support all the things that people care about with mucks. Jon: It was a little easier because we had a reputation. So a lot of people who knew video already knew us instead of oh, they're doing something new order. It. So that gave us a leg up. But it only takes you so far. And then, after the first few customers, you really do have to keep proving it yourself. [00:29:06] The area then how did you build trust and prove yourself within Amy: [00:29:10] the business? Yeah, in the beginning, that was that was quite hard. We did some open, some famous open source software. So that, that made it easier for some people to, to trust us. But in the beginning we had some companies would, 30, 40 million users trusting the core of their right. Thierry: To literally me and Tommaso, I was the tool. You wonder about their judgment doing that, but we're very grateful for them trusting stream seltzer early. And I think they [00:29:33] trust you, Amy: [00:29:34] I think. Yeah, our reputation and I think the pain was just more acute. Like they really didn't want to build this in house. Thierry: And a lot of them were smaller startups but also some really large ones. Yeah, and mean in general, we're just very grateful for the trust that those companies placed in us. But after you get the first few, you start to leverage that. So you are in conversations with some of the larger apps and you talk about your auto customers. Yeah. And nowadays you have these really large companies like click Hoppin Vimeo using using a stream. I started, we have some customers in common by the way, Jon. So that's and I think we try to signal it on the, on some quality metrics as well on our website. So nowadays we run chat on the edge, which is something that's a lot of work to do. But it showcases like how much we care about performance. We built our own storage technology on top of rocks to be and raft for some of the larger life event apps, because many life events, apps, these days use streets. We scale all the way up to 5 million concurrent connections on a single channel. So we do some of these things that, and we talk about that, which I think make a customer. Believe that there can trust a stream. And I think nowadays it's much easier than when we just got started. Also shameless plug we're hiring for iOS, Android go. Our core technology is Fritz and go and yeah, we may do other things that keep me up at night. Like just being able to hire and keep them [00:30:53] hire people. Yeah. That's interesting. Okay. So also one of the things that you just said that I want to pause and pull out there is that I think you're saying that to build trust and to get customers, you need to make sure that your product. And the problem that it's solving is more important to the customer than the fear of you fucking up their business. Amy: [00:31:20] Yeah. So basically in the beginning nine out of 10 potential customers will decide not to use you. Because of their particular pinpoints or because they're they want to do, they're more open to trying out new technology they started using you and then step-by-step you start to gather more customers and that it becomes much easier to convince a more established companies to these right guys as well. Thierry: [00:31:41] Okay. And I guess I also want to know what was the point in time of your business drawn that you realize that max was gonna be. Amy: [00:31:51] Yeah, that's that's a really good, interesting question because every stage is different. I would say. I would say from the beginning, we had a lot of conviction, but you never know was Jon: [00:32:01] there one big customer where you wrote just yup, this is it. Amy: I'm good. Yeah. [00:32:06] Pick a time. It'd probably be 2019. We started signing our first we launched our second product in 2018. So we have two products, both in video, second one, I think we started signing up some of the first, like really big customers off of that product. There are customers like Robin hood and Invesco who do a lot of video. Jon: And at that point it was like, okay, I think this is going to the right. [00:32:24] Yeah. Cool. That's cool. How did you Amy: [00:32:26] celebrate? Oh man. I'm terrible at that kind of thing personally. Cause every time something good happens. I'm like, that's awesome. Next. Jon: [00:32:35] Yeah. That's a very CEO mindset I feel. Amy: [00:32:38] But I th I think we did, we'll we'll do champagne for big deals and I was in the office when you know that kind of thing, maybe. I don't know. Jon: [00:32:46] Yeah. That's cool. The area, did you have a big moment where you were like, ah, yeah, this is it. Amy: [00:32:52] I think the, yeah, I think the last scary moment that we had for our first stream was when COVID just hit and quite a few of our customers started going, yeah. Thierry: So initially that looked scary, but then healthcare education, life events all picked up which we went from, oh, these numbers don't look so great. I hope this doesn't keep up to yeah. Things growing really fast. So that was really the last scary moment. But I think, as soon as you get to 50 K in MRR or. Which we did in like your year one or year two, you get to the point where you're a, as long as you're motivated and you like your team. There's usually a way to continue in to, to make the business succeed. And I also feel like we have a responsibility to our customers too. Cause there's been so many stories of API companies that shut down and didn't do well. I think our customers are a little bit scared of that as well. So from early on, we were very focused on making sure that our customers have a good experience in terms of uptime. In terms of like stability for the company. So we took yeah, we were very motivated to make that happen. And I think as soon as you can, Even low revenue numbers. At some point you have enough money to feed the team. And then then it's hard to become some, one of my investors said that companies are like unkillable off at a bunch and it's not always true, but I think to some extent there's some truth [00:34:06] to that, yeah. Okay. Okay. Cool. And let me peep into your corporate culture. How would you celebrate a big one? Amy: [00:34:18] are yeah. Similar to Jon. We are similar. We are not great at celebrating. We'll sometimes do some drinks with the investors, our role go for like a sports game or things like that, but we don't do a whole lot of celebrations. I'm always focused on the next target. Thierry: [00:34:32] Oh, I need to work. Jon: [00:34:36] I love a good celebration. So I'm going to take that opportunity every chance I get. Okay, cool. Amy: [00:34:46] I just I don't recommend that approach I take and I love when our team like takes a different approach because I think it's really healthy to celebrate it. Jon: [00:34:55] Yeah. Yeah. Especially because like you said, it, I think it's usually always okay. Amy: Onto the next like great job, everybody. Cool. Let's move on. And I feel like. I don't know, I guess this is some tips for you too, as a person who works as a startup in like a lower role, I feel like celebrating the small wins and be like building off the company's success makes me feel like a better employee. It makes me feel like more motivated to work harder. And then. Oh, you get the props or whatever, like people are recognizing you for your hard work. I think that's so important as a startup to, to help with yeah, like motivating people. I don't know. I think like it's such a beautiful part of my job that I love it. Okay. Cool. Thank you Q a guys very much for joining the podcast. I appreciated, I learned a lot about ABI is that I did not know before. So thank you very much. The area, if we want to find you and what we're, what you're working on online, where can we look? [00:35:53] So the website is get streamed thought. Thierry: I am [00:35:56] okay, Jon, where can we find you and what you're working on? Amy: [00:36:02] Yeah. Mux.com blog that mux.com. I write there a bit. Yeah, it was probably the best Jon: [00:36:08] three letter domain name. That's awesome. Thierry: [00:36:11] Yeah. Yeah, it was not easy, but we did it. Jon: [00:36:15] It's tight. Okay, cool. And then I also want to put a little note in here that these guys are nominated for startup of the year. Amy: So at the time of this recording, I don't know. Too much detail on where you can vote and what you need to do, but I will put that in the show notes. And so go, please go vote for these guys for a start-up of the year. Thank you very much. If you like this episode of the hacker noon podcast, don't forget to like it, share it, subscribe to it, do all the things. And this episode was hosted by me, Amy, Tom. It was produced by hacker noon edited by our lovely audio wizard. And you can find us at hacker noon on all the socials. See ya on the internet. Stay weird. Goodbye.