Ankit Kumar - CTO & Co-Founder at Ubiquity6

(Joel Beasley at 00:00:00) Hello, my friends. I'm curious to know how many of you have a leadership pipeline. We know that great leaders grow companies because we talk to them here on the show every day. But what are you doing to create great leaders within yours? If you're a CTO, it is 100% your responsibility to grow and improve your people beyond just their coding abilities.

(Joel Beasley at 00:00:20) We've built a tool that improves your people in their craft and in leadership. Visit leaderbits.io to learn more. Today, we are talking to Ankit Kumar, the CTO at Ubiquity6, and we discussed the advancements made in augmented reality, how the CTO role changes at various stages of growth, and tips to create a more productive working environment. All of this right here, right now on the Modern CTO podcast. Here we go.

(Joel Beasley at 00:00:51) This is the Modern CTO podcast. First of all, I love the name Ubiquity6. It sounds pretty cool.

(Ankit Kumar at 00:01:04) Yeah. Thank you.

(Joel Beasley at 00:01:06) I love what you're doing. Your website was beautiful. It was very clear. Edit reality together just sounded awesome. But how did it come about? How did you get started with this whole adventure?

(Ankit Kumar at 00:01:19) Yeah. So my cofounder and I, my cofounder, Anjney, and I have known each other for a long time since college. We lived together in college back in the day. And we always chatted about working together, building something. And I think the origin story of Ubiquity6 is maybe a little different in the sense that the vision of the company, what we're trying to build is not new.

(Ankit Kumar at 00:01:45) So it's not like we invented the idea. It's been around in sci-fi and sort of near future books. There's a book called Rainbow's End that is kind of an inspiration for both of us that we both read. So this world that we're trying to bring about has been thought of for a long time. And so it wasn't that we came up with an idea.

(Ankit Kumar at 00:02:03) It was more about when we were chatting, we realized that the tech is at a position. So to build such a world, there's a very multidisciplinary set of technology that you have to build and rely upon. And it was more of an understanding or a realization that across those different pieces of tech, each discipline is at a maturity now where the sort of sum of its parts is now possible. And so when we sort of got confidence in that and sort of did our investigation and talked it through, you know, would these things be possible? Could does this seem reasonable?

(Ankit Kumar at 00:02:41) We realized that this kind of world is now possible, and so we decided to build it. And it's, you know, I of course think mainly about the tech and talk mainly about the tech, but AR is also in a very good position from a consumer readiness perspective as it's been sort of introduced into Snapchat, Facebook. Consumers, generally speaking, have more and more understanding kind of what is AR, what should I expect from an AR experience. And so it was more of a realization that now is a pretty perfect time to actually build this world.

(Joel Beasley at 00:03:16) So what is it for the people listening?

(Ankit Kumar at 00:03:19) Yeah. I probably should have started with that. So kind of at a high level, what we're trying to do is turn physical spaces into containers for virtual content. So what we want to do is have a system where a developer or consumer can author content, edit reality as we would call it. And those edits or that content that they make actually reside in the physical world where it was placed by the creator.

(Ankit Kumar at 00:03:47) And so you can put any kind of content in any place in the world, and that is a multiplayer context that people can engage with and it sort of ideally conforms to this illusion that this is actually part of the world. And so the idea of Ubiquity6 is to build that virtual world and then let people inject their own content into it wherever they may be in their homes, their offices, the park across the street, museums, whatever it is. And so we're sort of that kind of virtual layer on top of the real world that is kind of persistent and multiplayer and sort of engaging.

(Joel Beasley at 00:04:25) So is it out right now? Can I install this on my device and begin to interact with this?

(Ankit Kumar at 00:04:31) It's not out right now. Hopefully, you'll be hearing things about that later on this year. We sort of, you know, we of course have our demos and previews. We did a preview at the San Francisco MoMA a while ago where we had a bunch of people get our app and, you know, not just people sort of in the company, but people just museum goers that were there and sort of interact with our experiences. And that was a really kind of nice event.

(Ankit Kumar at 00:04:56) And so we're sort of rolling it out, and we're testing it. We're making sure that when we do release it to the general public, it sort of just works.

(Joel Beasley at 00:05:05) I'm curious. Do you mind if I ask some deeper questions about this? I want to nerd it up in here. All right. So how do we keep the world from getting cluttered, man? And then how can I manipulate other people's stuff? How does that work?

(Ankit Kumar at 00:05:17) Yeah. So the sharing model and the privacy model from a sort of consumer perspective is very important. I mean, the privacy model in particular is very important. And what you're talking about is probably more in the sort of curation model than the sort of privacy model. And so we sort of view it as a couple of layers.

(Ankit Kumar at 00:05:38) Right? So we have our application, which is sort of the lens through which you are viewing our world. So when you open up our app, that is kind of the only way for you to see all this content, see all this sort of content that people place in the world. And that's sort of mediating the consumer and the virtual world. And we think of it maybe as a lens.

(Ankit Kumar at 00:05:59) And so that is where the logic of curation and sort of reducing clutter would happen. So, for example, the app layer might have a sort of following model where I'm following a certain number of creators. And so the world I see when I look through my camera is different than the world you see because you have different preferences. You are following the sports layer, and I'm following the basketball layer, whatever it is. I mean, you could be following content. You could be following people. A lot of those decisions are not necessarily set in stone for us. We're sort of trying to figure out what is the best way for people to interact with the world. You know, do you want to be curating your world based on people, who made the content? Do you want to be curating it based on the semantics of the content, like basketball versus football or whatever?

(Ankit Kumar at 00:06:45) And so those are decisions we're trying to make. And the nice thing is that it's sort of logically separate from the content itself. So you're right that if you had the view of all of the content that was ever made on the platform, it would probably be very, very cluttered. And so we're working on sort of how that curation mechanism happens, but it will probably be some kind of following of people that you care about or you like or content, something like that.

(Joel Beasley at 00:07:10) That's pretty neat. Because I get what you're saying. Right? So if I were to see everything, that'd be ridiculous. But instead, I'll have some sort of interaction with following creators, and I'll see their stuff. And then that way, it allows me to craft a world of my favorite things.

(Ankit Kumar at 00:07:27) Yeah. I mean, you can think of it similarly to a feed in a sense. So when I open up Twitter and you open up Twitter, we're seeing very different worlds, let's say. It's just that the display of the world is this vertical feed that we're going through. For us, the display of the world is the actual world, the physical area around you that you're seeing in AR through your phone.

(Ankit Kumar at 00:07:51) So it's a very similar concept to that.

(Joel Beasley at 00:07:53) That's actually really neat. I'm assuming you guys have your whole commercialization concepts all done. That's awesome. That's exciting. I like to see ideas of this because this is truly imagineering. Right? Like, what Disney used to do, your imagineers. I don't know if they still have them, but it's really forward thinking. Yeah. I love it. I just think it's absolutely brilliant.

(Joel Beasley at 00:08:17) And then your design and everything looks really, really clean. It looks very good, the brand and all of that. So I'm pumped up to try it. I want to try it. I want to do something in my office. I want to do something for the show and then have people look at. But here's where I also see it going. And I don't know. I'm just kind of on a random thought tangent here. But when the technology gets it to where it's in our contact lenses or our really thin, always wearable glasses, you know, because that's where the world is headed, whether we like it or not. You guys already have the platform and the content, and so the medium in which is right now, it's just the mediums in the app on the iPhone, but that doesn't mean it always has to be an app on the iPhone. It can translate to another medium. And now I could walk around the entire world and see my literal universe through whatever creators I'm following.

(Ankit Kumar at 00:09:04) Right. So I don't think it's that controversial to say that this virtual world that we're building is sort of best viewed through some ubiquitous wearable like glasses or contact lenses or whatever else. That is sort of probably the ideal way to view the content. We just think that that is far enough away and there's more than enough value in the mobile phone aspect that we're not really focusing on that, let's say. But you mentioned you have different potentially different sort of applications or views into the virtual world, and we sort of already have that concept.

(Ankit Kumar at 00:09:39) So, for example, you know, you'd have iOS and Android. So those are normally the same concept because they're both mobile phones, but they're different platforms. But we also have, for example, a web style view where you, as a creator, maybe can look at your virtual content in more of a sort of free flowing camera way and mutate it, change it, whatever it is. And that's just another view into our representation of the physical world. Right?

(Ankit Kumar at 00:10:05) And so even if I'm in San Francisco, if there's some content in Tokyo, I can kind of engage with people in Tokyo by way of browser based view that sort of teleports me there and gives a sort of representation of the world that's over there. And maybe some people who are looking at that representation are doing so in AR in Tokyo, but I'm doing so from my browser at home. Right?

(Joel Beasley at 00:10:27) I got it.

(Ankit Kumar at 00:10:27) And so it's just different ways to look at this representation of the world that we built.

(Joel Beasley at 00:10:33) Yeah. You gotta call the code base project name Teleport or something. That'd be really cool. Right?

(Ankit Kumar at 00:10:38) Yeah. It's called Atlas right now.

(Joel Beasley at 00:10:41) Oh, hey, dude. That's Atlas is equally cool. Okay. Where are you at right now with how large your team is?

(Ankit Kumar at 00:10:52) We are about 36 or 38, so something along those lines. We're growing.

(Joel Beasley at 00:10:58) Is that in engineering or the whole?

(Ankit Kumar at 00:10:59) No. That's the whole org. So engineering is something like 25 of that. We're still growing. So if anyone listening is looking to work on this kind of a project, reach out on our website or you can reach out directly to me.

(Ankit Kumar at 00:11:13) And yeah. So we're about 25 engineering, 38 or so total.

(Joel Beasley at 00:11:18) And so the website would be ubiquity6.com?

(Ankit Kumar at 00:11:21) Ubiquity6.com, or you can send an email to [email protected]. That's our recruiting email or reach out email.

(Joel Beasley at 00:11:32) Nice. It happens. I get emails from guests all the time. Like, hey. They heard my, usually, it's your story of who you are. They heard my story. They resonated with them, and then they moved across the country, and now they are some of my top engineers. And I was like, that's awesome, man. What a cool thing to happen through podcast. Right?

(Ankit Kumar at 00:11:48) Yeah. That'd be great.

(Joel Beasley at 00:11:50) So you're obviously your cofounder. You're there since the beginning. What sort of, let's bring some value to these people, right, that are listening. What sort of struggles did you have as you grew to 25, those first 25 people? What'd you learn?

(Ankit Kumar at 00:12:05) Yeah. So I'll speak on a couple of points. So one is sort of personally as CTO and kind of my personal growth along the way. And then the other would be kind of company wide. Okay.

(Ankit Kumar at 00:12:18) So from a personal perspective, I think anyone who does this kind of a thing where, you know, you're a CTO at a growing company, you should expect your role to change significantly as you grow. So early on, I was basically a coder. I was writing the main parts of the code base. And then as you get more and more people, it's obviously not that you become worse of a coder, but the way to sort of optimally exploit your knowledge of the system is more about making everyone else 10% better than using a 100% of your time to code. That's the way you best improve the company or the tech, let's say.

(Ankit Kumar at 00:13:04) And so some of the sort of struggles or hardships or learnings that I had to do early on is, and I don't think it's that uncommon, is learning the new, a different job. It's not the same job. Your job is changing, and so you have to learn the new job on the job. And that inevitably comes with some struggles and some, you just you won't be the best CTO for the time at your company, essentially, at any point in time because it's constantly changing. As we constantly kind of playing catch up about what is the best way to help the company at every time.

(Ankit Kumar at 00:13:47) And so you have to be in a mode where you're okay with that constantly changing, and that will inevitably lead to some frustrations. In terms of the company growth, I think that probably the main, the biggest, the highest value thing that I eventually learned was communication between teams, especially for us, for example, where because we're working on a pretty multidisciplinary kind of tech stack, we run the risk of having people in sort of silos working on different projects and very rarely understanding what other people are working on and how their project interacts with the ultimate product and the value that it's bringing. And so we had some troubles at various points around not communicating well enough and sort of not having the flow of information be baked into the system sort of, baked into the process that we went with. And I think that the biggest gains in productivity with respect to sort of challenges and growing and stuff has been around making sure that there's a good line of communication between all the engineering teams that so they understand where their part interacts with other people's part and interacts with the whole product.

(Joel Beasley at 00:15:07) And so what was one of the takeaways from that? Did you assign that responsibility to someone on the team to be constantly meeting with the other teams? How did you, on a tactical level, solve that?

(Ankit Kumar at 00:15:19) I mean, so for, we didn't, we don't have anyone explicitly assigned to that. At some level, I view that as part of my job because I will have conversations with, basically, all the team leads. And oftentimes, I'm the person who's saying or disseminating that information across the team leads who then disseminated down to the teams. I also think that the other thing that we've done is sort of, so we have a PM who is very technical and very sort of able to understand both the technical side and the product side.

(Ankit Kumar at 00:15:56) And I think one of the things that has helped a lot is empowering him a little more to be that middleman in a more technical capacity, not just a product capacity, to explain where different things interact with other things that primarily I would have explained to him or told him. And so maybe in that sense, there is a person who's doing it, but it's not so much like the job is to disseminate information. It just—we've sort of tried to put it into the structure of how we do planning and how we do stand-ups and sprint planning and stuff like that.

(Joel Beasley at 00:16:34) Yeah. It's hard. Like, this is not a super easy clear-cut thing. That's why we talk about it, right?

(Joel Beasley at 00:16:42) The more volume I could talk about this with other CTOs, the more feedback we can get. We're actually working on a guide right now that was inspired by Dailymotion.

(Ankit Kumar at 00:16:51) Mhmm.

(Joel Beasley at 00:16:52) Guillaume is their CTO over there, Dailymotion, and he said this phrase that inspired a guide, and he called it autonomous leadership. It was a centralized strategy with decentralized execution. And it was how they actually have all these teams running this huge stack, and they make a lot of the things themselves. They don't rely on a lot of the third-party tools. So it's, how did you get all those product teams? How do you get the central strategy with decentralized execution? And so we started collecting information from different guests about how they do that and putting it into a guide to share with other CTOs because I just thought it sounded really cool. I was like, that's a fun project in 2019, right? Autonomous leadership. I think that could be a buzzword.

(Ankit Kumar at 00:17:35) No, and I think that phrase is very true. That's kind of what I think most engineering orgs should go for is centralized strategy, decentralized execution. And I think that's the way for every different project to move the quickest.

(Joel Beasley at 00:17:52) And what you described too with how you had this PM who was there and capable—what I felt like you described was you letting go of something you were doing, which is good. It's the hard thing to do. It's you recognize this individual is capable in the right spot, so you have to back your responsibility out. And that's so tough because you start to fall in love with what you're doing and it's exciting and you're like, oh, I'm really good at it. And the moment you get good at it, and the moment it's exciting, it's like you have to let it go and then go do something hard and difficult again.

(Ankit Kumar at 00:18:18) Well, that's really the life of a CTO, a co-founder. Like, as the company grows, you're constantly letting go of things and moving on to the next thing and changing your job to make yourself most productive for the company.

(Joel Beasley at 00:18:37) Yeah. We're doing a company over here. So we did the podcast, and that created this leadership program. But, you know, so far, I've had to go into sales and learn how to do sales and then build a sales team. And now we have two sales teams that run autonomously, and we get sales without me. And it's like, in seven months learning how to do sales, and I had to, and then I had to go do that with accounting, right?

(Ankit Kumar at 00:19:02) Right.

(Joel Beasley at 00:19:02) That place was crazy. I was like, let's hire somebody in accounting real fast.

(Ankit Kumar at 00:19:06) Right.

(Joel Beasley at 00:19:07) And payroll and all that stuff. So yeah. But it's definitely hard because I got good at it. Like, I started closing some sales. I'm like, this is awesome. Like, I love it. And then I'm like, oh, now I gotta go focus on product because we need to improve our product now. And I'm in the middle of that too. And it's like, every four, six weeks, you're learning something, dropping something. And then when I'm dropping something, I'm basically putting a person in place to run the thing that I figured out. It's like, alright. I figured this. Now do this.

(Ankit Kumar at 00:19:39) It's basically learn something so that you can get someone else to do it, and you can learn something else.

(Joel Beasley at 00:19:47) Yep. And then I'd learn all the people and leadership skills with it. It's like you're doing a leadership company, learning leadership skills. Like, one thing I found is that when people will research stuff and then come back to me and give me a bunch of feedback—I just found this recently, so I'll share it with you—like, not to give feedback in line because it almost, like, if a third party were in the room, they'd say, oh, it's almost taking the wind out of the point. Like, when they're giving me feedback, my job is make them feel heard. And then after, I process their feedback, come up with some useful stuff. Because I noticed if I start doing the real-time feedback, I get engineering. Like, my engineer part of me kicks in, and I'm just like, that won't work because this, this, that, and I start firing off too fast. And then I'm like, nope. I'm getting loads of feedback from people. I just look at them in the eye, pay attention to the feedback, and then process it, come up with some useful stuff, decide how I wanna deliver that back to them. And that's work. Like, it's work. It's leadership.

(Ankit Kumar at 00:20:50) That's, and that's also a skill that you have to learn. That's probably in terms of all the skills that you learn as a CTO or co-founder that you learn and then drop, that's a skill where that's always your job, essentially. Like, that is always part of your job no matter how big you get, at least up until now anyways. And my guess is that, no matter how big you get, those skills are probably the most ultimately valuable.

(Joel Beasley at 00:21:15) Yeah. The communication. So from entry-level CTO to NASA and Microsoft, the trend there over the past year is communication, right? Be super aware of your habits, and then it's all about the people. And those are the ones that tend to just continuously come up. It's like when you're programming and you hear the phrase, like, everything's an object, but then it, like, as you learn more and more, that phrase becomes more truthier, right? You're just like, wow. Everything really is—everything becomes really demystified and simplified as you go up to the levels. And the same thing is happening on my beginning journey of this leadership stuff. It's like, it's about people, it's communication, it's your habits, it's how you spend your time. So let's bring some tips real quick. You curate environments, essentially, in the digital and physical world. But you yourself, like, what tips do you have for your environment that you work in for your productivity? Have you added things, removed things from your environment to become more productive?

(Ankit Kumar at 00:22:16) Have I added things or removed things from my environment to be productive? Well, I will say that it's a really good idea to get a really good chair and desk and monitor and keyboard and mouse. Like, that sounds obvious, but I feel like I will often see people working just on their laptops still, like their MacBook Pro or whatever to code. And I feel like the simple tools of your job are actually super valuable. So, like, get a standing desk that you can go up and down, get a nice chair, get nice keyboard and mouse and stuff. Because that stuff, you use so much throughout the day that even a minor improvement is so valuable. So that is something I will definitely say. Other than that, it's really just whatever your preference is. Like, you know, I always have some kind of drink on my desk, and I pace around a lot. I actually have a little Dave and Buster's ball at the office that I kick around as I walk around, and that just helps me think. And so if you have anything like that, then, you know, don't feel weird in doing it, I guess. Whatever floats your boat, I guess.

(Joel Beasley at 00:23:37) I'm a mover. I'm a pacer. I'm a wanderer. Yeah. When I do my talks and stuff, I have a kitchen, you know, at my house, and it's like a center island thing, so I can walk around the kitchen in the room. And so I just, I walk in these circles. My wife calls it the track, right? Because I walk in the circles, and I'm typing, and I'm saying my talk stuff in my head. And that helps me so much.

(Ankit Kumar at 00:24:05) Yeah. Me too.

(Joel Beasley at 00:24:06) Yeah. Doing some meaningless tasks, like putting together a chair or kicking a ball around, like it gives the—there's something that happens with your brain that's very useful.

(Ankit Kumar at 00:24:15) Yeah. Our office is similar, actually. It's a big loop, basically.

(Joel Beasley at 00:24:20) Okay.

(Ankit Kumar at 00:24:20) And so I'll just kind of walk around it, kicking this ball around, thinking.

(Joel Beasley at 00:24:24) Now where is your office located?

(Ankit Kumar at 00:24:27) On 3rd and Market in San Francisco, downtown San Francisco.

(Joel Beasley at 00:24:30) Oh, I know that. I'm there probably, like, I don't know, four or five times a year.

(Ankit Kumar at 00:24:35) You should, next time you're in, you should come get a demo.

(Joel Beasley at 00:24:38) Oh, for sure. That'd be really cool. There's a lot of great companies out there and great people. I mean, I think some really cool leadership people out there with Williams-Sonoma. They just bought an AR company.

(Ankit Kumar at 00:24:52) Oh, yeah?

(Joel Beasley at 00:24:53) Yeah. They had, like, 110 people in it, and they bought it because they were, I guess they were, you know, digitizing their stuff for the future, right? You're a kitchen company. You wanna see what the kitchen piece looks like on your countertop.

(Ankit Kumar at 00:25:06) Yeah. Yeah. Right? That was one of the early—I mean, IKEA had done a demo app as well. That was one of the early sort of promises of AR. One of the early use cases that people kind of latched onto in a sense.

(Joel Beasley at 00:25:20) Yeah. What do you think of it?

(Ankit Kumar at 00:25:22) I think it's a good use case. Yeah. It's less—I would say it's less the use case that we go after in the sense that it's not really about persistence, it's not really about building a sort of persistent world. It's more about using AR as a tool to understand something, which is a perfectly good use case for AR, and one that would exist perfectly fine on our platform as well. But we care more about the use cases where you're sort of building something persistent that you're gonna keep coming back to and interacting with.

(Joel Beasley at 00:25:56) Now are you giving talks this—do you go out and give talks about what your company is doing and your technical?

(Ankit Kumar at 00:26:02) Sometimes. I've done different podcasts as well, and I've given talks at various conferences and stuff like that. I would expect over the next, you know, year or two as we get closer to launch and have something out there to do more of that. There's a certain amount of, sort of, like, almost education in a sense because AR is so new and this medium is so new. And the AR things that people are familiar with, which is getting more and more—they are getting more and more, you know, out there—but those are typically still in the realm of local AR. So, for example, capture a video of it on Snapchat or something like that. And we're trying to take that next step into persistent AR that actually you keep coming back to over and over in physical space around the world. And that is a pretty significant shift, I think. So there's a certain amount of education in a sense.

(Joel Beasley at 00:26:59) Yeah. And that's important too. It's like, when I looked at your company and I saw you guys had raised some money, I was like, that's pretty cool because I think whether you ultimately achieve your goal, what you will invent and come up with in the process of attempting to achieve it will become very valuable, right?

(Ankit Kumar at 00:27:21) I hope so.

(Joel Beasley at 00:27:22) Like with NASA when they, you know, we have power tools because of the space race, right? Because you needed to be able to use tools in space, right? Like, that's where tech transfer, when stuff comes out of the government and into the private sector. So I see a lot of, you know, I think a lot of the early AI-type startups, they got acquired because someone saw, oh, you built that technology. You were trying to do A, but I needed to use it to do B. And then they pick them up. I see your company as having a high potential for that scenario. I don't know why, but it just comes up in the back of my mind.

(Ankit Kumar at 00:27:57) Well, it does—I think maybe one of the reasons you think that is because when you think of the sort of foundations that we're building, there are so many different ways that could be used. Because, you know, the way I framed it earlier in terms of turning physical spaces into containers for virtual content, like, that content can be anything. In a sense that, you know, today, the main container for content is 2D screens. So your mobile phone, your browser, your desktop, whatever, your monitor. And, fundamentally, we're trying to introduce a new medium for content. And it can be any content, but it's just 3D and spatial and persistent instead. And so, yeah, I think that the number of applications for the tech that we're building, the foundations that we're building are very broad.

(Joel Beasley at 00:28:51) I love it. What are some of the products that you love that are helping you accomplish this mission?

(Ankit Kumar at 00:28:59) You mean products in terms of inspiration or that we're actually using?

(Joel Beasley at 00:29:03) Like, things that you—in my mind, I was thinking of things like Slack and GitHub.

(Ankit Kumar at 00:29:07) Yeah. Yeah. Yeah. Yeah. So Slack and GitHub are two really good examples. GitHub Actions, which recently came out, we don't really use yet, but I really like what GitHub is doing. So Slack, GitHub, we use Jira and some Atlassian tools as well. I think that you need some tool to keep track of progress and planning and stuff like that. Of course. What else do we use though? I mean, we use a lot of AWS, and there are a lot of places where AWS, people have their gripes about it and, you know, perfectly fair. But overall, it's such a nice thing to have that this concept exists where they have managed services that do these very difficult-to-build things essentially out of the box. So I think that in general, when people are building things from scratch, you should try to leverage as much as you can without having to do it yourself. And so there's nothing wrong with that in a sense. So, AWS overall.

(Joel Beasley at 00:30:24) Do you know what Actions were? Because I just learned about that from you just now. GitHub Actions.

(Ankit Kumar at 00:30:29) Actions? Yeah. They're kind of like a serverless execution model that GitHub is hosting for you. And so, like, it will trigger some action. You can set up actions to trigger on certain GitHub events. And I think they're containerized. I'm not exactly sure what the definition language is, but it's like a serverless—

(Joel Beasley at 00:30:53) It's like workflow automation. Yeah. I looked up—I cheated. I looked up their website. Yeah. It's workflow.

(Ankit Kumar at 00:31:01) They recently announced it. There are a lot of people talking about it, and I think it's a very promising direction for GitHub as a company, as a product.

(Joel Beasley at 00:31:13) Yeah. Because right now, I think I handle this through Heroku. Like, Heroku does a lot of this. But—

(Ankit Kumar at 00:31:18) People do that. People use third-party stuff like CircleCI and stuff like that for the CI. But it's CI/CD-related actions.

(Joel Beasley at 00:31:26) Yeah. We use—

(Ankit Kumar at 00:31:27) But I would imagine that GitHub is gonna expand into things, into promoting uses like that, like testing and even deployment and stuff like that.

(Joel Beasley at 00:31:36) Yeah. We had the CTO of, or the creator of Jenkins. Remember, have you used that before?

(Ankit Kumar at 00:31:41) In a past life, yeah.

(Joel Beasley at 00:31:43) I used it like once or twice.

(Ankit Kumar at 00:31:46) That actually brings up a product that I really like that we use called BuildKite.

(Joel Beasley at 00:31:51) Oh, I haven't heard of this.

(Ankit Kumar at 00:31:53) Yeah. So BuildKite, it's an interesting thing. It's a CI/CD thing, and it replaced our usage of CircleCI.

(Joel Beasley at 00:32:00) Really?

(Ankit Kumar at 00:32:01) And so the model is that they run the orchestration piece. So you install kind of—they call them BuildKite agents—on your machine. So you own the hardware, and they're stateful machines. We use these two boxes. So they provide the agent and the orchestration, but you provide the hardware for CI/CD jobs.

(Ankit Kumar at 00:32:23) So when you put a PR on GitHub or something, it'll trigger—the triggering and the allocation of the agent to run the job is done by BuildKite, but the actual compute is provided by you. Whereas in, for example, CircleCI, you give a document in there and then they run the container, right? And the reason that was so good for us is another tool that we use that I really like, but sometimes actually has some frustrations, is Bazel, if you've heard of that.

(Joel Beasley at 00:32:54) I haven't used it yet.

(Ankit Kumar at 00:32:56) Bazel is Google's open source version of their internal build system called Blaze. And it is basically like a cross-language build system where you define sort of the dependency graph of your code and how to build various things, and then it will do the caching and the recompilation stuff for you. And it's very good for monorepos, which is what we use. We have all of our code in one repo. So you can sort of have very fine-grained dependencies and explicitly stated between different packages.

(Ankit Kumar at 00:33:27) So you can sort of separate your code much better. And so the problem with that—is not the problem with that, but Bazel is much better if you can keep it warm. It's like a persistent server model where there's a cache of its analysis of your code base. And so there's a cache in terms of the compilation and everything as well. And the problem with the pure container orchestration things is that every time a new container gets allocated, there's no cache anymore. It's potentially a new box. So we really wanted our CI/CD, which is run through Bazel, to be warm when a pull request happens. So then we want to own the box, and so it's a stateful CI/CD set of machines. But we didn't want to deal with the orchestration of allocating the job to things ourselves. So BuildKite does that basically perfectly for us.

(Joel Beasley at 00:34:21) Oh, that's awesome.

(Ankit Kumar at 00:34:22) That's a great tool. If you're using Bazel or other build systems that need warm caches, BuildKite is definitely something to look into.

(Joel Beasley at 00:34:30) That's fantastic. I will—I'm actually gonna take a look at that. I think that's pretty neat. There's also some issues. Sometimes you get some false positives from CircleCI and things like that.

(Ankit Kumar at 00:34:39) Yeah. I mean, that will happen either way probably, but yeah.

(Joel Beasley at 00:34:43) True. All right. So I've learned an absolute ton from you. One last question I have is, have you picked up on how your team will pick up on your energy? Your personality is if you're down, it mirrors across your network. Have you picked up on that? And then how have you adjusted things in your life to make sure that you're positive?

(Ankit Kumar at 00:35:08) You know, it's funny you should mention that. I think that I typically have a pretty neutral demeanor all the time. I'm not very often explicitly positive or negative. And so I don't think I really run into that issue. I'm sure if I come in and I'm super, super negative, then people will feel that, and I'm sure they'll respond appropriately or correspondingly. But I don't—I'm actually pretty neutral, just always, I think. So I haven't really noticed that. Maybe I'm not picking up on something, but I haven't really had that particular problem.

(Joel Beasley at 00:35:51) Yeah. So you haven't noticed that they'll pick up on your mood?

(Ankit Kumar at 00:35:56) Well, I'm saying that I don't really have too much difference in my mood.

(Joel Beasley at 00:36:00) Okay. So you're saying your neutrality—

(Ankit Kumar at 00:36:03) I'm pretty neutral.

(Joel Beasley at 00:36:04) You're pretty neutral. Is your team pretty neutral?

(Ankit Kumar at 00:36:07) No. Not everyone. I mean, everyone has their own personality. Yeah. My cofounder is not that neutral. He's very positive—

(Joel Beasley at 00:36:14) Is he?

(Ankit Kumar at 00:36:15) Often. And I think that makes me feel positive. And then I'm kind of neutral in general.

(Joel Beasley at 00:36:20) So you don't go up on the positivity train? You recalibrate?

(Ankit Kumar at 00:36:24) I don't go down on the negativity train either.

(Joel Beasley at 00:36:27) So you're super centered?

(Ankit Kumar at 00:36:29) I'm pretty centered, I think. Yeah.

(Joel Beasley at 00:36:31) And then that way people can gravitate around you, right?

(Ankit Kumar at 00:36:35) Yeah. I mean, that's—

(Joel Beasley at 00:36:36) I like it. I like—it's so interesting too, because one thing that I have found to be true is that just because it works for one person does not mean it works for other people, right?

(Ankit Kumar at 00:36:46) Right. Right.

(Joel Beasley at 00:36:47) So I'm—that's why I like having, you know, being the nerd that I am. I like having this wide array of data from different leaders because it's just so useful to understand the different types and styles of leadership.

(Ankit Kumar at 00:37:01) Yeah. That's actually something—there's something related that someone told me early on in this company, is that there's sort of different archetypes of CTOs, generally speaking. And across all of them, probably 70% of the job is the same. You gotta be a good planner. You gotta be a good people person. You gotta sort of be a good allocator of technical resources, decision maker in terms of what we're gonna invest in. And then the rest—30%—is sort of up to you to define what you want that to be. You can go into full manager area where it's more people person stuff, and you're focusing pretty much primarily on making sure people are getting what they want out of the company and stuff like that. You could code still. You could sort of do investigations into new technologies. There's different archetypes. And it—there's not a one-size-fits-all. You have to pick what's best for your company, but probably more appropriately, pick what you are gonna be best at. What makes sense for you as a person.

(Joel Beasley at 00:38:13) So where did your strengths fall? What did you do with that extra 30%?

(Ankit Kumar at 00:38:17) I probably fall on the side of investigating new technology or different approaches. Things that are very hard to schedule as part of the day-to-day planning of doing the work that you need to do because they're kind of investigative or speculative, and it's unclear if it'll work or if it's valuable. And so people need to—it's valuable to sort of do that investigation off the books, so to speak, and then maybe integrate or maybe not. So I would probably fall more on the technical side. So I'll be coding things or investigating things, which typically involve some proof of concept stuff.

(Joel Beasley at 00:38:56) So what I've seen as the companies hit the several hundred people is with your—with that style that you described—is essentially an office of the CTO or office of the CIO forms, and—

(Ankit Kumar at 00:39:07) It's like the labs—

(Joel Beasley at 00:39:08) Where you will have three people you keep close and you do lab things, things that the other—and you'll solve either—do some research from outside the business or look to solve problems in an interesting way that are happening inside the business, right? So I always think that that's cool. Whenever there's an office of the CTO, I'm like, oh, they're solving some crazy black box, hidden problems. I want to know what they're doing. They can never—

(Ankit Kumar at 00:39:31) Yeah.

(Joel Beasley at 00:39:31) Know. I—

(Ankit Kumar at 00:39:33) I feel like that's probably the mode that we'll grow into. I mean, we're small enough now that the office is basically me in that sense.

(Joel Beasley at 00:39:39) But—yeah. That's how it starts.

(Ankit Kumar at 00:39:42) But that model doesn't work for everyone, right? And doesn't work for every company. And so it's really an effort to understand what is best for you.

(Joel Beasley at 00:39:53) Yeah. I found out—I would have told you I was the labs person probably five years ago. But going through this recently, I realized that I'm the speaker person.

(Ankit Kumar at 00:40:04) Mm-hmm. You're right. Yeah.

(Joel Beasley at 00:40:05) I like to go out and talk and talk to crowds. I like—that's what energizes me. And it's interesting because if you were to talk to me, you know, five, six years ago, that same energy came from me solving ridiculously hard technological challenges. But then for some reason, I adjusted, right?

(Ankit Kumar at 00:40:26) Right.

(Joel Beasley at 00:40:26) I changed, and then I get that energy from helping large groups of people in person. I like that. That's a lot of fun.

(Ankit Kumar at 00:40:35) Yeah. Yeah. And it can change—I mean, it changes too. So who knows?

(Joel Beasley at 00:40:39) Who knows? It might go back to labs. I don't know.

(Ankit Kumar at 00:40:42) Right. Yeah. Yeah.

(Joel Beasley at 00:40:43) Make some, like, organic—I've always had, like—I'm always curious to know, at what point will our devices start—will we have an adapter for our biological power that we generate to run small devices? Someone's gonna do it at some point. I think we already do it in medicine to some degree, but we're energy powerhouses, man. We generate ridiculous amounts of energy, right? We just convert it through food, caloric intake.

(Ankit Kumar at 00:41:07) Right.

(Joel Beasley at 00:41:08) So at some point, these small little devices that we all have on our wrists, at some point, someone's gonna figure out a way to tap power from our organic structure more than just—

(Ankit Kumar at 00:41:19) Oh, yeah.

(Joel Beasley at 00:41:20) Yeah, right?

(Ankit Kumar at 00:41:20) Yeah.

(Joel Beasley at 00:41:21) It's like, I gotta eat two more hamburgers today because I'm wearing my power device.

(Ankit Kumar at 00:41:28) My iPhone's running out of battery.

(Joel Beasley at 00:41:29) My iPhone's right out of battery. Yeah.

(Ankit Kumar at 00:41:31) Yeah.

(Joel Beasley at 00:41:31) More celery sticks. So if you were to go back to yourself, like, the day you decided with your awesome, optimistic cofounder to pursue this venture. If you could tell yourself one piece of advice that day, what would it be?

(Ankit Kumar at 00:41:51) Well, first of all, I would say do it. Okay. There was a decent amount of—we did a decent amount of sort of investigation, and we wanted to make sure that this was possible, let's say. And so I would say, like, do it. It's possible kind of thing. And then I guess—I think I would say—I think that my what I thought it would be and what it was, what I thought this journey would be and what it has been were very different. That it's not worse or better, but it's just different. And I don't think the thing I would—I don't think I would give the advice—I don't think I would say, like, this is what the journey will be. I think the advice I would give is just expect that the journey will not be what you think it is. And be ready for big changes and sort of working on uncertainty and stuff like that.

(Joel Beasley at 00:42:58) I love it, man. It's fantastic advice. For me, it would scare the hell out of me. I would just be like, it's awesome. You're gonna love it. You're gonna do great, buddy. You're like, it's big and it's not what you think and it's a little bit scary, but keep going.

(Ankit Kumar at 00:43:15) Well, I said do it. I said do it.

(Joel Beasley at 00:43:15) Yeah. You said do it. Dude, you're a lot of fun. I like you a lot. When next time I'm out in San Francisco, I'm gonna stop by and say hello.

(Ankit Kumar at 00:43:22) Yeah. You should. And check it out. You get a demo too.

(Joel Beasley at 00:43:26) Yeah. I'd love to see your demo.

(Ankit Kumar at 00:43:26) Yeah.

(Joel Beasley at 00:43:26) So people want to find out more about you. Are you more active on Twitter or LinkedIn?

(Ankit Kumar at 00:43:32) I'm not super active on either. Twitter is probably the better way.

(Joel Beasley at 00:43:36) Okay. So they want to tweet at you. They just look you up.

(Ankit Kumar at 00:43:39) Yeah. It's Ankit Kumar SF.

(Joel Beasley at 00:43:42) Okay. That—

(Ankit Kumar at 00:43:42) That's my handle.

(Joel Beasley at 00:43:44) Very cool. And then—yeah. You gave the email earlier if they're interested in a career—

(Ankit Kumar at 00:43:48) Yeah.

(Joel Beasley at 00:43:49) At your company.

(Ankit Kumar at 00:43:49) Adventure@Ubiquity6. Yep.

(Joel Beasley at 00:43:51) [email protected]?

(Ankit Kumar at 00:43:54) Yep.

(Joel Beasley at 00:43:54) Awesome, man. Anything else you want to put out there into the world?

(Ankit Kumar at 00:43:58) No. This has been a lot of fun. Thanks for having me.

(Joel Beasley at 00:44:01) Awesome. Yeah. Thank you for coming on. You have a fantastic day.

(Ankit Kumar at 00:44:04) You too. See you.

(Joel Beasley at 00:44:11) Thank you so much for listening. If you'd like to help, please take a moment right now to open up the iTunes app and leave a review of the podcast. If you take a screenshot of the review and text it or email it to a friend who needs to listen to the podcast and then CC me, [email protected]—if you CC me on the email, I'll send you a copy of the Modern CTO book or give you a shout out on the podcast, whichever you prefer.