Tejas Kumar

Transcript from Friday January 27th, 2023


Carl Vitullo: Tejas, thanks so much for joining us. Excited to have you. I know you have worked in developer relations for years. You've worked at companies like Vercel, Xata, G2i, Spotify, and you're now working independently as an educator and content creator. [00:00:13]

Yeah. Anything else? Any major points that I missed? [00:00:16]

Tejas Kumar: No, you got it man. That's that's a great introduction. I like how short it was. It's it's nice. Now we can get into actually a nice conversation. [00:00:24]

Carl Vitullo: heck yeah. Yeah. Maybe we can start off with, I know you are currently starting a company and it's a little bit too early to get into the fine details of it, but Yeah. Why are you starting a company.

Why are you starting a company [00:00:35]

Tejas Kumar: That's an excellent question. And I found, I, I really don't do it lightly because I know that nine out of 10 companies fail and all of this. But I feel like the stars kind of align for one, I've wanted to do this for years. And two, I, I don't know if y'all have had Sunil Pai on this on this like discussion office hours thing. [00:00:54]

Carl Vitullo: We haven't actually. I'm familiar with him and he's an incredible person. I followed him for years. He's great. [00:00:59]

Tejas Kumar: Yes. I highly recommend. Let's, we need to make that happen because he's, especially now, he's doing the same thing, right? He's. He started a company called Cool Computer Club. He's building something awesome called Party Kit. And he took me aside in November in a tech conference after I had quit my job at Xata. [00:01:15]

And he said, listen man you are in a very special point of your life right now because this is one of the rare occasions where you'll have literally no job and all the time to do all the things you've ever dreamed of. Also I've worked for the last 15 years nonstop. I've been an, I've been an employee for the past 15 years without a break. [00:01:32]

And in those 15 years you come up with dreams and things and you hear, more importantly, you hear about the pain that companies have. And my most recent role in Xata as the director of developer relations, I got to see the pain that a lot of companies have to endure doing developer relations. [00:01:50]

And then I became an angel investor last year and, got to see like from the cap table perspective how hard DevRel is for companies, firstly to hire great people, and either there's not enough talent or there's not enough talent that the companies are willing to pay for. You do get great talent at a very high price. [00:02:07]

And so companies will end up like maybe hiring like less than that price. And they might hire inexperienced people and then it's, DevRel is so young that it hasn't really been codified very well. And so what happens is either the person doesn't perform according to the company's expectations and they let them go. Or the company uses the vague role of DevRel to overwork people and overwhelm them, and both ways. One of the sides, if not both, has trouble. So I've seen this pain and the company I'm starting aims to codify and fix what we think of DevRel and really help companies get the most out of DevRel. So I'm really excited about that. [00:02:46]

Carl Vitullo: Yeah, that makes a lot of sense. I think of DevRel as being, it's one of those things that I think now a lot of teams, especially companies that run like SaaS products or otherwise have developers as one of their target customers think of it as something that, they need to do in order to develop their business. [00:03:04]

But I think the finer points of "what does it mean to start a developer relations program" are not well understood. So is that sort of the niche you're look, you're looking to explore? [00:03:14]

Tejas Kumar: Exactly not just looking to explore, but looking to surf. Is this exactly, that is companies one thinking about do you even need DevRel and two if the consensus that you do, which we're seeing by the way, just a massive like hockey stick style growth of dev tools companies these days as well. [00:03:30]

DevRel and community are immeasurable [00:03:30]

Tejas Kumar: And chances are most of them need DevRel. But then, how do you go about doing it? A lot of people have no idea how to measure DevRel. In fact, you'll see a lot of conference talks, even some by me, where people, they try to answer the question, how do you measure dev? And the answer they come up with is just don't because they say it's immeasurable. [00:03:49]

Carl Vitullo: Yeah that's something that I feel, deep in my heart and soul about. How I've thought about community is, there, there are a lot of metrics I could track in running Reactiflux, and I, there are metrics that I do track, like number of active members week to week or whatever, that's information that I use to inform I don't use it as a target. [00:04:09]

It's not oh no we lost 15 active members. We need to bump those numbers up. And, too often people try to come up with metrics and then those metrics become the only thing that matters. Yeah I definitely I relate to what you said about just don't do it, don't measure. As soon as you start measuring that, those become the target that becomes, reason to do it. And the reason to do it is so much broader than those metrics. [00:04:34]

Tejas Kumar: Exactly. And it's so interesting you say that because the word relations isn't the name dev relations. Imagine if you took your relationships with your friends or your close, family even and said what are the goals for this relationship? How do we, that's immediately a turnoff in most relationships because it, it ends up like commoditizing the people. [00:04:56]

And I've, this is one of the mistakes I've seen companies make is they, through whatever wacky DevRel measurement process they'll try to come up with, they turn people into commodities or assets to be exploited. And that deep there's a lot to be said about devaluing people with that, but I think that's a big criti that's a big factor. [00:05:16]

And I think the solution to it is to recognize, okay, this is something that has a lot to do with people. And in many ways it's, it has parallels. I would even say with politics. Like a politician's approval rating is very similar to a dev tools company's approval rating among developers. And so you have to go at it from that campaigning angle. [00:05:34]

Like you, you don't see like presidential candidates just outright, saying, Hey, what's the, like what's the goal of campaigning? They'll probably have those conversations behind doors, but they're trying to really connect with people on a very real level. And I think that's where DevRel stands. [00:05:51]

Extractive relationships in DevRel and community [00:05:58]

Carl Vitullo: Yeah, that makes a lot of sense. And it, something that I've said and thought about a lot is in, in community, and I think also within developer relations, one of the things you can optimize for is engagement. Who's reacting, who's participating. But as soon as you start trying, as soon as that becomes the goal, it's it incentivizes you to do a lot of things that I think are harmful to the broader picture of it, to what like you said, to forming relationships with these people. [00:06:21]

Like I think one of the common approaches to build a community when companies try to do it, is just to do things like giveaways or, raffles. and it just sets up this like transactional exploitative nature of this relationship, which is just it just poisons the well, and as soon as you start from that place, it's so hard to claw your way out and get to authentically and deeply connecting with people. [00:06:47]

Swag and broken trust [00:06:50]

Tejas Kumar: Yeah, exactly. This is where, so at Xata, I'm really proud of this man at Xata. When, where I was leading DevRel. We had the best reputation on any conference sponsor, booth floor. And that's because we had a lot of swag to give people. And it was also the most inclusive swag. Like we had earrings, we had like t-shirts in both men and women's cuts, for example. [00:07:10]

And people showed up and it was really interesting. They were like hesitant to take the stuff we just openly wanted to give away. And when we asked them why they said they, they were just surprised that they could take it. They were like they said to us, oh no, these other companies with booths they asked us for contact information in exchange for this water bottle. [00:07:26]

And they looked at us in disbelief. They're like, wait, so we can just have it really? And they're like, yeah, can I have two? Sure. Because we don't we spend the money. We got these things. We don't really care about, extracting money and contact information from you. We want to genuinely give you this because we want to bless you. [00:07:41]

And if you either use our thing or you don't, that's secondary. We can talk about that later. So this, what does this do? Already sets the groundwork for a healthy relationship. It's look unconditionally, we're gonna support you. And if you want to, if you don't want to support us, that's fine, but that builds way more trust than something transactional like you were talking about. [00:08:00]

So thinking through these really high level things and working with companies to train their DevRel people to think at this level is what my new company's about. [00:08:10]

Carl Vitullo: Very nice. Yeah. Okay. I like that a lot. Yeah. And it's even just hearing you tell that story, like those people, those companies and groups who approach the relationships from that sort of transactional nature, I'll give you this if you give me that. Like it doesn't just poison their own well, it sets a standard that people then come to expect from all companies doing that. So yeah I could see how there's a lot of value in getting people to do it well, like not just for those companies, but for the whole industry. [00:08:40]

Tejas Kumar: Exactly. And can you imagine an industry where all of the dev tools companies compete with each other on who can best serve their developer communities and who can best bless their developer communities. What's awesome about this? Through that competition, developers win, a lot. And that's what I'm going for is what's best for developers and developer communities at the cost of companies that can afford it. [00:09:02]

Carl Vitullo: Yeah. Okay. I like that a lot. [00:09:04]

Transactional relationships [00:09:07]

Tejas Kumar: With Reactiflux, right? Like the, I'm sure like talking to you about this, you've got the same passion, I would say for developer communities and giveaways and wellbeing. I guess I don't see anything transactional in Reactiflux almost ever and what I do see is a lot of, service to the community. [00:09:21]

So it sounds like you would have resonating thoughts about that. [00:09:25]

Carl Vitullo: Yeah. Trying to not be transactional is one of the reasons I try and do events like this nobody, no money is changing hands here. We don't have any sponsorship for these events. We just, I just try to reach out to people who I think will have interesting things to say and who I can have a good conversation with and benefit the members who are interested in tuning in. [00:09:47]

It's interesting to hear you say that you think of this as a very non-transactional space, because I actually think of it differently. I think that one of the struggles we've had at a, community level is that 95% of the members, the only way they participate is by asking for and providing help, which I think is a very transactional relationship. [00:10:08]

It's, you have a problem, you hop into, Reactiflux, you ask, and in a couple of hours you've had a conversation. Cool. Thanks. Bye. Trying to break away from that sort of hop in, hop out, transactional help is, has been one of the things that I've been, I don't wanna force a change, but just trying to provide opportunities for deeper relationships to form and doing events like this, I think is one way that we can encourage that, so this is we do some office hours. [00:10:38]

I'd like to do more office hours where we get community members just talking about their experiences. As well as, I thought I've had recently as refocusing events where we have external guests coming in like you and position those as a "community spotlight." Here is somebody who we, I as a community leader think is doing something really great and deserves a little bit more recognition. [00:11:03]

Tejas Kumar: Wow. Thanks. It's a nice compliment in there. I appreciate that. [00:11:08]

Carl Vitullo: Yeah. You said in, as we were preparing that one of the reasons you were think you were interested in doing a company was to not be one of those people who keeps talking about one of these days doing something. [00:11:21]

Can you talk a little bit about that? [00:11:23]

Fear about trying something new [00:11:28]

Tejas Kumar: yeah. I've been a part of this discord for a while though I haven't really asked for help or anything. But I feel like everyone here, the 80 plus people that are here have the ability, through code, to build literally anything we want, any software we want, any app we want, right? [00:11:39]

I'm sure there's in the room, over a hundred people in the room. There's potential and ideas like, oh, I wanna build this. Oh, I've been dreaming about this one app and this, but we don't do it. And I've fallen afraid of that myself, where I working in DevRel for as long as I have, I've seen so many pain points and so many things that I think could be improved. [00:11:58]

And I spent the past year 2022 working in this space and kind of just complaining oh, if I had a company, I would do this differently. Oh, I would what was I'd hire all the women and people of color and fix the state of, inequity and all this stuff. And it's, I find it, what I was saying earlier, it's very easy to like shake your fist at the sky and say these things. [00:12:16]

But when I thought about actually doing it multiple times, I immediately felt fear. And this fear immediately made me back away from it without actually backing away from it. It made me, and this is something I don't know I'd love to hear if you can relate. It made me back away from it. But still at the same time feel like I'm doing something towards it. [00:12:35]

It's so weird, Carl. What I mean by that is I quit my job, right? And I thought, as I mentioned, Sunil Pai told me, "hey, you're in a great time to do that thing you've been putting off." So I said, great, I'm gonna do it. But then I'd start to feel the fear and then I'd be like, you know what? No, I'm just gonna take a job. [00:12:51]

I'd call CloudFlare. I'd be like, Hey, I heard you're hiring developer advocates. Can I and that would happen over and over again to the point where I got really frustrated with myself. Why do I keep applying for jobs I have this dream? And ultimately it took some reflection when I realized, listen It's better to try this thing, to build it, maybe fail than to not try at all. [00:13:14]

And that statement is really what cemented me to start building this thing. And you know what? It probably will fail Carl. Nine out of 10 startups fail. And that's, I don't think that's a bad thing. I feel like it's, it gives you closure. When I was a younger single guy there was this whole thing about oh, let's see if I can talk to this pretty girl at the thing. [00:13:31]

And a lot of guys were afraid of rejection. But like on the other side of that, Quote, unquote, failure at the time when you're a young man chasing women is oftentimes a feeling of Hey, wait a second. I did that and I tried and now I'm not gonna spend the rest of the day regretting that I chickened out. [00:13:49]

So that's what's really the driving force here. And I feel like there's people here who are similar, who feel like an inkling of entrepreneurialism, but it's snuffed out by fear and also by people around them telling them, oh, maybe, pay the bills with this or that. [00:14:03]

And and full disclosure, like it's very risky and I'm watching, my family finances go down day by day every day that my company doesn't make money. But it's part of it. It's part of it because I feel like in tech we have this bubble where like we just take our money and, react front end engineer salaries for granted and everything's. [00:14:25]

Given, you know what I'm saying? And to be able to be able to feel like human things once in a while. It's very exciting. Anyway, sorry I'm ranting a bit, but hope that answers your question. [00:14:35]

Carl Vitullo: Yeah. No, I think that was a great answer that I there's so much to respond to there, to in there. I really deeply and viscerally feel what you were saying about the fear of starting a company… actually taking a real shot at it is so hard because, it probably will fail. Most people who try something new do not replicate their software developer salary in the first year or even in the first five years. [00:14:58]

Tejas Kumar: Yeah. There's the other side to it too. There's the other side where you expect to fall, but then you fall in reverse AKA flying. And this happened to I don't know if y'all have had James Q Quick on the channel. [00:15:10]

Carl Vitullo: We have not, I'm familiar with his name. [00:15:12]

Tejas Kumar: He was, he's been a developer advocate for most of his career I'd say. And he was working at Planet Scale, so a direct competitor to Xata at the time. That's how I know him. And he got laid off, dude. And he got laid off and he, again, he was in the same place I was without a job, and he was like, I'm and when he got laid off, I was working at a competitor and I messaged him instantly and I said, oh, you got laid off Mr. Competitor, come work for us. We'll treat you good, right? And he said, yeah, cool, cool. I'll get back to you. But first I've been dreaming of being a YouTuber for such a long time. [00:15:44]

I wanna just try this YouTube thing. I wanna be a full-time content creator. See how it goes. We'll probably fail, but whatever. Carl, like within the first year being a content creator, he made basically his salary back just doing it. Exactly. And that's, those are the stories we need to hear more of because I think, we hear too much of the negative and it's probably good cuz it protects us, but there's also the other side. [00:16:06]

Carl Vitullo: Yeah. Yeah. I do. I've been listening to a podcast, People I Mostly Admire, which I would recommend. It's really great. It's from one of the authors of Freakonomics. And one of the points that he hits on in a lot of his interviews is, "talk more about failure." Just because fail, it's so common and there's so much you can learn about things that don't work. [00:16:29]

Like, why didn't it work? Was this, a, what are the specific details about this attempt that made it not work? And really just normalize that yeah, you know what, you try something new, you try something hard, you try something out of your experience. And yeah, it probably won't go well but you learn something, you get better at it and you can try again later. [00:16:49]

Tejas Kumar: Exactly man. And that learning it's for one, like from, so I'm currently living that, Carl, I'm in the middle of that, right? And it's exhilarating. It's like learning how to deal with dude. I wrote an SOW myself the other day, like a statement of work. [00:17:04]

Carl Vitullo: Yeah I'm in the same boat. I'm writing those too. [00:17:07]

Trying something new; an exercise in empathy [00:17:13]

Tejas Kumar: Dude, it's nuts. spent three hours writing a 13 page contract thinking, oh my gosh, I better hire someone don't know, it's just exhilarating not knowing where your next paycheck's gonna come and working I'm reminded so I grew up in a Muslim country and I'm reminded of, I, if there's any Muslims in the audience, but I'm reminded of, Ramadan from the Muslim it's this thing where you for an entire And the reason they do it, they They don't eat like from sunrise to sunset, which in some countries closer to the poles where a day is 14, 16 hours, it's pretty wild. And they do this for a month. and I was wondering why when I was younger and they said it's because we want to feel what people who don't have a lot feel, it's literally, it's a month long exercise in empathy. And that I think is pretty cool. [00:17:50]

And that's exactly where I think we both are because it's not to say we're I've been an employee in a startup. I've been employee number six. I've been an employee at a large company like Spotify. I've been an employee at a mid-size startup like G2i. [00:18:02]

What I've had in common with all these roles is I've worked with founders, but I've never been a founder and I've never been able to empathize with the things they have to deal with. Then this is what I'm talking about, like an exercise in empathy that way. it's nuts, man. I will probably never speak ill of another founder ever again. [00:18:19]

Carl's past failed company [00:18:25]

Carl Vitullo: Yeah it's a hard life. It's, it is just full of so many challenges. I'm in a similar sort of spot. I've been working independently for the past year trying to do community and technical community consulting and yeah, it's, this is not my first company that I founded, but going back to what you said about feeling the fear to get started the last time I started a company was like 2014 or 15, and I was trying to do a subscription box service with my partner at the time. [00:18:50]

Tejas Kumar: Nice. [00:18:50]

Carl Vitullo: like we did it for about a year and a half and ended up shutting it down and just I learned so many things through doing it badly. Oops, I didn't file all my taxes right. here's a two grand in IRS… scary letters from the IRS are not fun. But, I, so I did it badly once and then I spent six years working at companies and not trying again. [00:19:11]

But then that experience of having done it so badly and failed pretty hard just gave me, when I started thinking about it again and okay, I wanna take, I wanna do it seriously this time. What does that mean? It just gave me, it shoved the door open, just that little bit much wider where I understood more about what it takes and how to do it right. [00:19:30]

Tejas Kumar: you're, you're in a similar place now. I, you may or may not be starting something, but you're, you've been independent Do you feel like you might lean towards putting that, into putting those things you've learned in practice again, [00:19:42]

Carl Vitullo: Yeah. I am doing that a little bit right now. I, but more in the consulting work than building. I don't know. I think of consulting very differently from building a startup, building a company. Yeah. But another dimension of it that I'm just really enjoying is the freedom to pursue self-directed goals. [00:20:00]

I think that is just such a great aspect of working independently or trying to start a company is needing to take stock of short-term goals. Okay, where does the next dollar come from? What is work that I can do right now that will get me closer to that next dollar, but also set up something bigger down the line? [00:20:19]

Tejas Kumar: yeah. So you think short and long term at the same time? [00:20:22]

Carl Vitullo: Right. It's so challenging though. [00:20:23]

Tejas Kumar: Yeah. No, I agree. I, what you've done for me, Carl, in this discussion already, which I'm super thankful for, given me confidence that failure isn't that bad. I'm full disclosure, man, I'm terrified, but like talking about how you failed with this tax stuff and you took a job and it was fine, and then you tried again. That's It's a bounce back story. [00:20:41]

Carl Vitullo: Yeah, And I mean ultimately we work, we all work in tech. We have a lot of skills and I think that is, so freeing in a way that like, yeah, there's always something to fall back to if doing something on your own doesn't work. There's always other jobs. [00:20:53]

Tejas Kumar: Yeah. And I feel like that's a really good encouragement to the people listening. Is there's always, especially in a community like this, there's a job board here in Reactiflux, you know what Carl, that's where I'm if everything fails and I need a job straight to the job board. [00:21:06]

Carl Vitullo: Straight to the job board. There we go. I didn't really expect to get so deep into entrepreneurship and whatever on this, but I'm glad we did. [00:21:12]

Tejas Kumar: Me neither. I do feel like it's, I don't know if it's interesting for the people, I mean with the name like Reactiflux, I expect us to be talking about react and things which not complaining. I could talk about entrepreneurship all day. Especially cuz it's currently my whole personality and adventure. [00:21:28]

Carl Vitullo: Independent content creator. But yeah. Let's bring it back a little bit more into the technical side of things.

New web technologies you're excited for [00:21:34]

Carl Vitullo: What are some new technologies that you're excited about in web development? In the web space? [00:21:40]

Tejas Kumar: Ooooh. [00:21:40]

Carl Vitullo: Big question. [00:21:41]

Tejas Kumar: That's an excellent question. That's a great question. So part of my independence from employment is starting a YouTube channel. And this answers that question in a broad context, but I'm not gonna be one oh, just click the link and go to my YouTube type of answers. [00:21:55]

A lot of people have asked me actually about this YouTube channel, like, why are you starting it bro? Theo and there's Primagen, there's all these YouTubers already. [00:22:03]

Carl Vitullo: Crowded space. [00:22:04]

Tejas Kumar: It's a crowded space, but I feel what I'm, what my intention for YouTube is not really to create tutorials. [00:22:09]

There's plenty of coding tutorials and they're awesome, but my intention is to like, talk about tech that I find in the web and not just talk about like we just did with the entrepreneurial stuff. Like deeply review them and discuss their trade offs, and what's exciting and what's worth it and why it's exciting and so on. [00:22:26]

Zod and tRPC, type safety on network calls [00:22:27]

Tejas Kumar: To bring it back to your question, one of the videos I'm currently working on is a video on, Zod, have you heard of and used Zod Carl? [00:22:34]

Carl Vitullo: Oh, I've heard of it. Is that's the like validation library, sort of like, Yup or Joi [00:22:39]

Tejas Kumar: It's like that, but more, and that's why I love it so much. It does do validation of types and ske and json. If it's an object, you'll do validation on schema as well. But it also pairs perfectly with TypeScript such that you don't have to write types and TypeScript anymore. [00:22:54]

You just do type of whatever your Zod thing is and you get a TypeScript type just for free. [00:22:59]

Carl Vitullo: Wait that's pretty cool. [00:23:00]

Tejas Kumar: it completely changes way you, it works as a source for your types, which then you take and plug into type native things like tRPC, to generate like APIs and things. [00:23:10]

And then what you've end up, what you've ended up creating is a completely end-to-end type safe client server relationship, but also at runtime, not just build and compile time because Zod validates on the run time. Your apps are just way more robust with Zod and tRPC. [00:23:27]

So I think those two are technologies in the front end space that I'm, and in the backend space because they're I'm quite excited about those as full stack holistic TypeScript technologies. [00:23:36]

Carl Vitullo: Yeah, definitely. Yeah. I think that tRPC was the context in which I had have heard about Zod most. Yeah. And I've been hearing for years about generating types off. API schemas, I think the first person heard talk about that was Jared Palmer back in like 2017. He was, generated a GraphQL schema and then used that to create types for the front end consuming it. [00:24:01]

And I just remember it was like a light bulb, like mind explosion moment. Like, wait, you can do that. Holy shit. I can, prevent my app from shipping a breaking change. What!? So yeah it's cool to see new approaches to that. [00:24:14]

Chronological Snobbery and jQuery [00:24:18]

Tejas Kumar: Yeah. I also the code gen route is actually, gosh, I'm, so there's a term one of my favorite authors CS Lewis uses, he calls it chronological snobbery . That's when we look at the past and go like, oh my gosh, that's so primitive. You know? Some of us would look at jQuery and talk about it like that. [00:24:32]

Oh my gosh, jQuery. But jQuery still runs most of the web. It runs more of the web than React But what I'm about to say sounds very chronologically snobbish. But, this like Graph QL code gen type stuff is no longer the premier way, no-breaking-change, type-safety, API server and client like it, it just isn't, because… [00:24:53]

think about it, right? Like you, you have your terminal talk to some server, do an introspection query, and then generate files on your file system. Those files can be out of sync with the server and you have to work on some type of like Chokidar, like watcher magic to make sure that when you save the code is gen. It's clunky because there's so many moving parts. Whereas tRPC, is all of it. tRPC does the schema description exactly like GraphQL and it also exports for you a data fetching library that's type safe using your description and has runtime validation with Zod. It's taken the best of this graph ql code gen stuff, and like evolved it to 2023. [00:25:35]

It's quite exciting. In fact, again, I have to thank you Carl for this discussion because it's basically writing my next YouTube script for me. [00:25:42]

Carl Vitullo: Great. Yeah, no I, tRPC is definitely, I haven't I haven't yet had occasion to use it, but it's definitely top of my list of tools to reach for. [00:25:52]

React as a middle aged man [00:26:02]

Tejas Kumar: Yeah. tRPC and Zod. About what's exciting also, I think there was an awesome tweet, man. Have y'all had, I'm about to ask you if you've had another person on this Feel free to feel free just to know, but have y'all had, Jani Eväkallio on this, on, on [00:26:05]

Carl Vitullo: Oh no we haven't. I'm familiar with him though. He worked at Formiddable Labs for a long time and is, I think he's also doing an entrepreneurship thing right now. [00:26:13]

Tejas Kumar: Yeah. He's got a, AI writing tool, which is phenomenal. I bring him up because he tweeted something today that was hilarious. I thought it was so good that I retweeted it. He said React is in its middle-aged man era. It's at the top of its economic productivity, but culturally completely irrelevant. [00:26:32]

At the top of its economic productivity, but culturally completely irrelevant. I thought that was hilarious. Because React is exactly that. It's not maybe completely irrelevant, it's just this like mature thing that cool, it exists, but it's not really as sexy as it once was. [00:26:49]

Carl Vitullo: Right, it's not the, hype generator that it was, five years ago, five, 10 years ago. [00:26:54]

Tejas Kumar: Yeah, exactly, because like server components is this huge shift and it's a massive one, but people don't seem to care as much as they did about hooks. [00:27:03]

Carl Vitullo: Yeah. That reminds me of years ago when I was at, working at my first startup, I read, I, I can't remember if it was like a TechCrunch article or just like a blog post, but it used the metaphor of a clock face to talk about startup life cycle. [00:27:19]

And it really, it put this idea in my head that I've seen apply to all sorts of things, and I think that's a good example of it, where, early in something's life. It is, it's the darling, it's the, everyone loves that. Everyone's so excited for the possibilities. And then over time as those possibilities get explored, and some of them pan out, some of them don't, the energy drops and like skepticism comes in and then like later on as it gets older and starts to fade out of relevance or hype, then people start cheering its downfall. And I don't think react is quite into the cheering its downfall phase of that, but it's certainly into the the phase of its life where it's getting a lot more skepticism than it used to. [00:27:59]

Tejas Kumar: Yeah, but I think part of that comes from the same reason why there is so much hate on JavaScript. And Ja for example, like I could literally open my console right now and type for you JavaScript type of null. And the type of null is an object, right? And oh my gosh, that's so stupid. [00:28:15]

React compared with Qwik's design goals [00:28:27]

How can null be an object? I And JavaScript will still have some hate, but there is no arguing with the fact that it is the best language, the only language of the web so I feel like React is there, React is locked into its own design and unfortunately, It's design is outdated. [00:28:36]

It's design does not allow for some of the sexy new stuff we're in libraries like Qwik and Solid, because it was not designed for that. And you have things like quick coming in this ridiculous level of code splitting where a pool of workers can in unison, collaboratively hydrate, not even hydrate collaboratively, send different parts of your website to the browser pa in parallel. [00:29:01]

Carl Vitullo: Oh, interesting. [00:29:02]

Tejas Kumar: Nuts And yeah, it's absolutely nuts. And react can't even come to this zip code because it was never designed for that level of code splitting and distributed So I think that's losing its sex appeal cuz there's plenty of sex appeal being had in newer things that are able to be designed for modern architectures, [00:29:19]

Carl Vitullo: yeah. That's interesting. I haven't, I've heard of quick, but I haven't really dug into it before. So that's a new description to me. So it, it uses cloudflare's edge workers, that sort of technology. [00:29:31]

Tejas Kumar: It can use that. You know react, when you ship, react to a user. What you're shipping is React js, the, that's the browser, like React library. You're shipping React dom, which is massive, which is like 200 kilobytes. And if it's a NextJS site, you're probably also shipping like the NextJS run time. And if you use get static props, you're probably shipping like a huge bunch of json. You're shipping a lot like the JavaScript you'll multiple kilobytes, sometimes hundreds. [00:29:55]

Qwik is one kilobyte constantly, always on first load, actually less. Because it's so small, they inline all of Qwik directly into the HTML that you serve. And then how it works is when you hover over something interactive, or rather, when you interact with something interactive, let's say for example, you have a button that has an on click callback. At build time quick will wrap that callback with the dynamic import to the file that contains the callback. [00:30:24]

So you write code that looks just like react, but at build time Qwik will split out closures like from on click handlers into separate modules. It'll replace your on click handler with a dynamic import of getting that on click handler. So then your user, enjoys the one kilobyte first load clicks on a button, which triggers a dynamic import of this other module. And that then pulls that down the wire and then your call back. The page rather the JavaScript bundle slowly grows as people use your thing as opposed to everything upfront. And the benefit is if a user never interacts with something, they never download the code for it. [00:31:04]

Carl Vitullo: Yeah that's I'm thinking of optimizations I've done to add code splitting on, by a page or by a feature or something like that. That was all a manual work. I had to figure out where the seams were and like, and analyze like what was coming and where to figure out where a good seam to introduce was. [00:31:20]

Okay, so quick does all of that– [00:31:22]

Tejas Kumar: Automatically. [00:31:23]

So when you're talking about finding the seams, right, Carl? I feel a big deterrent to code splitting and React is, closures. In the body of a function component, if you have use state at the top, then In the component's return in the JSX tree you have on click and you have something that uses that state variable from outside the scope of your function. You can't code split that because it's using something outside its closure. So Qwik does that, it does like compiler level magic to put the closure and the outside state together in some file that is lazy loaded And it does all of that for you. And it just like splits literally everything. [00:32:04]

Carl Vitullo: Okay. That's very cool. Yeah, cuz one of the sort of, I think of it as a failure mode for optimizations I've tried to do with code splitting. It ends up where, okay cool, great. The code for this particular page is loaded asynchronously, but the only code that is only used on that page is like four kilobytes. [00:32:25]

Okay, great. You're, you've cut four kilobytes off your initial load, but now you've introduced more latency when people try to get to that page. So it's a wash. [00:32:32]

Tejas Kumar: That's why I'm so excited about Quick also, by the I'm not like here trying to evangelize quick, we kind of just drifted at the topic, but I feel like, builder io, so the people who make quick, this is their like, full-time job to think about all this stuff that we tried to think about and I've pressed them hard. I've spoken to the guy in charge and I was like, okay, dude, but I get that one kilobyte first load and my, my internet connection dies. Then I can't even click on your button cuz it's all dynamic, it's all lazy, loaded. And he looks at me and he's like, no dude. And I said, explain. [00:33:00]

So when it loads that first kilobyte it also has a manifest of things to prefetch. And as soon as you get pixels on the screen and that first paint, it immediately starts prefetching all of the other modules. Not only does it prefetch all of the code split modules in the background, so not blocking it also prefetch them in an order of likelihood to be interacted with that it calculates. [00:33:26]

Carl Vitullo: Do you know if that's like part of the runtime, is it will do, like realtime monitoring of what people are interacting with and update that priority list. They do. [00:33:34]

Tejas Kumar: think, I think it is right now, I think that's why it's beta, because that's pretty hard to get right. I [00:33:39]

Carl Vitullo: Yeah, that's a hard problem. That's interesting, huh. Okay. That's very cool. [00:33:43]

Tejas Kumar: Yeah. Exactly. And since everything is code split, I think the really cool thing is you could serve different parts from different workers all at the edge and get these, like of trademark, blazing fast load times. And so when you have all of this innovation in the UI space and you look back at React, it's this tweet from Yani about React being this middle-aged man that's not so relevant, but at the top of its productivity, I think is just on the nose, like on hammer on the nail. [00:34:12]

Carl Vitullo: Another thing I'd love to talk with you about is, I know you're working on a course platform, is that right? You and Sara Vieira. [00:34:22]

Tejas Kumar: Sarah and I are building this thing because we want to empower people to make great stuff and profit from it. And also we want to empower the people who download and buy these courses to enrich their careers. I feel like in some way it's a direct competitor to egghead. [00:34:39]

But one thing about me, Carl, I, I hate competition. I I'm the most anti-competitive person on earth. it's not like I hate the people that compete against me. That's not what I'm saying. I hate the notion of competition. I think it's a scarcity thought. [00:34:52]

And I'm more instead of saying, we compete, I prefer to say we complete the landscape. And so we do the course platform is basically egghead, but in many ways, less curated. Like If you've tried to teach on egghead, they'll send you like a microphone and they have like guidelines for how you can do a course I feel like they should, because there's so many instructors. We're going about it by being a bit different. Instead of. Opening the net really wide to everybody initially being an instructor and then policing them with, you know, use this microphone and speak this way. [00:35:27]

We're just gonna be pretty casual about it. Hey, you want to teach? Here you go. You have a basic webcam and mic. Awesome. Good luck. And give them the tools and the training and whatever they need as teachers ourselves. And then say, good luck. Sell your course for whatever you want. Use whatever you want. We'll support you. And also an to try and empower people who aren't well represented in the industry as Like women and people of color, like if they wanna teach giving them the best skills and tools to do that, and creating a platform on which they can sell their courses. [00:35:58]

I've heard from a lot of women actually who want to do courses that. Like on YouTube, for example, like, oh, I would love to teach on YouTube, but I'm scared of the comments I'll get from people. [00:36:09]

Carl Vitullo: Oh, yeah. [00:36:09]

Tejas Kumar: So creating a platform where that's and allows people to feel empowered instead of intimidated is something that's important to us. [00:36:17]

Carl Vitullo: I've seen a lot of independent educators building their own their own platforms. You I'm thinking of like Matt Pocock and, Josh Comeau, both developing their own educational platform rather than going to somebody like Egghead. So, is is, is that the sort of creator you're thinking of serving more, or is it a little bit more similar to Eggheads [00:36:39]

Tejas Kumar: No, these people are famous. also Pocock is not doing his own platform. Is teaching. He's partnering with Badass Courses. [00:36:47]

But yeah, no, Josh Comeau is, in my opinion, Carl, Josh Comeau is like the best educator I know. He's up there with Kent Dodds. I respect Josh immensely because of exactly that. Like, he'll build the platform. He puts so much care into it. Like he makes these like awesome 3D graphics go. It's just beautiful. Like nobody can disagree with the fact that it's just stunning work. [00:37:11]

Not necessarily Josh as well, because Josh has, the passion, the drive, the skills, the everything [00:37:18]

Carl Vitullo: He'll build everything he needs. [00:37:20]

Tejas Kumar: We're after the people who, feel like they could but are maybe not fully confident in themselves. We're after the kind of people who, like if you go to them and say, I believe you can sell a bunch of courses cuz you're very smart. They would look at you and get shy and be like, oh, I don't, I doubt that I'm skeptic. [00:37:37]

So people who like doubt the people you want to help sell courses. And the cool thing is we're building it using, react NextJS. We're doing it in public over live streams every so often. We had one today, I'm exhausted. It was a like three hour live stream. But yeah, we're doing it in public to show people listen, like it's, we want to, do it in public to, to show people how much we mess up and hate the fact that sometimes doing auth is hard. [00:38:02]

Carl Vitullo: Right. Yeah. That's funny. I feel like authentication is one of those things, everyone needs it, so it can't be that hard, right? No, no. It, it's hard. [00:38:10]

Tejas Kumar: Yeah. And oftentimes it's because of the silliest things. It's like, Oh, I forgotten my dot or I put something in my dot N, but I forgot to restart the dev server or something like this. [00:38:19]

Carl Vitullo: Classic. Okay. Yeah. Very nice. I love the, uh, the build in public ethos. I think that makes a lot of sense for a course platform as well. [00:38:27]

Tejas Kumar: Yeah, [00:38:27]

Carl Vitullo: it's kind of dog food to your own process. Show what producing the content looks [00:38:33]

Tejas Kumar: exactly. And the cool thing is once we're done with that, the platform's just gonna be open If anybody wants to take it and use it for their own, like Matt Pocock TypeScript course, whatever, they can just feel free, like clone the repo, do whatever you want. So the plan is to monetize the actual course content from creators. [00:38:49]

And of course creators keep 100% of whatever money their course makes. What we're thinking of doing is every creator course creator, like they plug in their like stripe to the platform. And then, it's just a direct line between the person on the other side who clicks buy and the creator. We, we don't do any app store, like 30% nonsense. [00:39:10]

All of it, just for the sake of education and learning and fun. I'm sure you're, you can relate Carl. It's another way of expressing I am unemployed and I have way too much time, [00:39:21]

Carl Vitullo: Right. When you have so much time, it becomes a challenge to fill it. Yeah. Well, you've, you've definitely got a lot of things going on. [00:39:29]

Tejas Kumar: Yeah, I'm really excited about it. it. And then, this chat has so many insights and ideas that I didn't even know I had. that's because of your awesome skills as a host and conversationalist, so I appreciate that. [00:39:40]

Carl Vitullo: That's the whole point of talking to people is see what you didn't know you knew. [00:39:44]

Anything else you wanna touch on before we wrap up? I'm curious to hear what your experience with Angel investing has been. If you want to share a little bit about that [00:39:53]

Tejas Kumar: yeah. As a first time non-employee founder person, I've, I've been diving a lot into finance actually, and investments and taxes and all of this. And, I see more of the value of Angel investing actually now as a founder. it's been good. it's been, It's been interesting. There's criteria to, to be an angel investor, oftentimes you can even circumvent that if you really want to. But these safeguards are put place to protect you because angel investing is like the riskiest type of investing on Earth. And it has to be, it has to be money that you're 100% comfortable losing. [00:40:27]

It has to be cash that you're willing to set on fire, like the Joker in The Dark Night. [00:40:31]

Carl Vitullo: Are you referencing like the accredited investor check? [00:40:34]

Tejas Kumar: yeah. So that check is also very interesting cause it doesn't help those of us who don't live in the usa. So I'm more than capable of, at least when I started, I was more than of angel investing. Because I was talked into it by my friend, Show Depot from Polygon. [00:40:47]

He is like, dude, write some checks. And I was like, sure man. How? [00:40:49]

Carl Vitullo: How? Big question. [00:40:51]

Tejas Kumar: Anyway, long story short. Well it turns out it's not hard. You talk to founders and you say, Hey, I wanna support with an angel check. You them your address and your money and your ID stuff, and they have their lawyers look at it and they send you, a contract saying you'll pay them that much. [00:41:06]

And most often it depends on the type of company. Some company will straight up give you like ownership contractually in exchange for your angel check, of course proportionate to your angel check and the valuation of the company at the time. Angel checks vary in size, and I've seen the smallest angel checks are like $5,000. The largest angel checks are $800,000. I've seen weird sizes of all shapes. I wanna be the kind of who's comfortable enough to Angel invest $800,000, honestly. [00:41:38]

But proportionate to your investment, you get, an ownership stake in the company. Some companies do this new thing that Y Combinator invented called A SAFE, which is a sale agreement of future equity. it aims to just simplify it. So you give them money and they give you an agreement to sell you equity in the future, when the valuation is higher. And you usually will have some type of like maximum valuation cap. [00:42:02]

Carl Vitullo: Right. Okay. Yeah. Right. That's like a convertible debt, I think is a term used to reference that? [00:42:06]

Tejas Kumar: Yes, it's the, it's like convertible debt version two is the SAFE. And yeah, that's basically it. It's just you sign contracts, you send addresses, you send money. And over time, the hope is that the companies have some type of exit. Either they go public, they sell, or some other very wealthy investors who's up and they're like, oh, what are those shares? Who does that belong to? I'll buy it from them. [00:42:27]

Carl Vitullo: Ahh, secondary market. [00:42:28]

Tejas Kumar: Right. And they have to buy it from you at the valuation of the company, which And then you get way wealthier. But and all the angel investing I've done, it's literally money I don't mind this money that I didn't even know I had. [00:42:41]

Dude, I've been a tech employee for 15 years and over that time I've accumulated scary amounts of money, as I'm sure multiple engineers have. And I didn't know what to do with it. So I decided let's just angel invest and what I've angel invested, I'm okay to lose, but I've strategically angel invested not so much in the companies, but in the founders. [00:43:02]

In the founders who I'm like, listen, like if their company fails, there's no hope for any of like some of the smartest people in the world, who I'm convinced are able to pivot and save their company in an emergency event. But who I'm also convinced that will not have emergencies cuz they're just absolutely brilliant So for me it's about the people and then whatever company it is, I don't really care. But then, there's also some savings that I was not comfortable betting with angel investing. And I learned about like safer investments like ETFs. There's a whole bunch I could say about all of that. [00:43:34]

I've, since I've become unemployed and since I've started, I recently invested 60% of all the cash I've ever made my entire life in different places. This happened like three days ago. And dude, like losing quote unquote, losing 60% of everything you've ever worked for, is pretty nuts. [00:43:51]

And I did it for multiple reasons. One, to beat inflation because at least in Germany, year on year, cash is losing 10% of its value year on year. So I did it to beat inflation, but also I did it to feel the squeeze. I wanna feel like, oh my gosh, we don't have any money. the hope is it would motivate me with this business be successful. [00:44:11]

Carl Vitullo: Yeah, everything you're saying, just I feel like I lived at some point over the last year. It's just funny to hear someone else doing the same things as me. Probably six or eight months ago I had a similar kind of… [00:44:23]

I've been working in tech for 10 years, I've made an amount of money, and over the last year, especially going independent, it's like, okay, one source of money is income through selling a product, selling my skills as a consultant. But you know, what, another source of income or a source of money is, growth in investments. So I started thinking of it as well, no, if I'm gonna work for myself, maybe one of the highest leverage actions I can take is smartly investing the money that I already have. [00:44:57]

Yeah, so, over the past year I ended up speaking to, a bunch of advisors and reading books and articles and about like diversification and tax loss harvesting and all the financial details and with the amount of money I have invested, like if I can get an extra of 4% gains through diversification or whatever, then that could, that could extend my runway of how long I can be independent by months or years even. [00:45:25]

So, it's, it's one of those things working independently and needing to proactively think about where your money will be coming from rather than just, semi passively taking a paycheck every two weeks. Is it really forces you to think differently about where your priorities are. [00:45:40]

Tejas Kumar: Exactly, and I feel like as engineers we're all capable of doing that. Cause we spend most of our careers optimizing one thing or the other. So we could optimize that, [00:45:47]

Carl Vitullo: Yeah, yeah. We spend our whole careers learning new skills and putting them to practice immediately. well, why can't one of those skills be financial literacy? [00:45:55]

Tejas Kumar: Exactly. [00:45:56]

Carl Vitullo: All right, well, I think that's a good place to end it. Tejas, thanks so much for coming out. This was isn't a great conversation. It went so many [00:46:01]

Tejas Kumar: I really appreciate it. Thanks for having me, Carl. [00:46:04]

Carl Vitullo: Yeah, thanks so much for coming out. [00:46:05]

Tejas Kumar: All right. Take care. Bye.