Algora | Algora TV

Aiden Bai, founder of Million (YC W24) - Open source interviews #36

Algora

Watching now

00:00 Hey everyone, welcome to the Open Source Founder Podcast.

00:04 Joining us today is Aydan Bay, creator of Million.js, the open source project, and the new company behind it.

00:12 Aydan, thank you so much for joining. I'm super excited to do this and would love to start with some information about your personal background and then how Million got started.

00:21 Yeah, thanks for having me on. I'm a big fan of the podcast big fan of uh the bounty program

00:26 and all that thank you all right let's kick it off are you you're 19 years old and currently in

00:34 college or around yeah i uh i actually just dropped out uh in january so 19 just popped out and then now i'm living the dream phenomenal wow and how long how

00:48 many years has a million been going for around three years now that is crazy so what like 15 16

00:56 years old living with your parents and kind of like releasing this piece on the internet right

01:00 exactly yeah it's a lot of a lot of interesting things just like working through

01:06 school and then just working on the side project and suddenly blowing up and just a lot of fun

01:11 fun stuff very fun very unique and congratulations i think recently you also went through kids round

01:19 the open source accelerator by joseph jacks Richelson also helping. Tell us about that.

01:26 Yeah, Gitround is amazing.

01:28 I kind of applied on a whim.

01:30 Actually, Peter was like, either JJ or Peter was like, yo, check this out.

01:34 I was like, okay, I'll look at it.

01:36 And I was just like randomly applied.

01:38 JJ was like, you're a great fit for this program.

01:40 And I was like, great, amazing.

01:42 It was really cool seeing other commercial open

01:46 source companies like be in the batch and also be able to interact with them and call up jj

01:52 basically anytime or peer anytime and just ask for advice which is like super valuable um overall

01:58 get round is really cool you should apply for batch too and yeah great yeah i think these are the very best people in this industry to get

02:06 advice from and so congratulations for that and then there's also y combinator right yeah and then

02:13 then i was like you know what for like december i was like you know what what if i applied to yc

02:19 now i'm here so what's what's that like the the program is running at the moment right so what's

02:27 going on yeah this is great like it's really cool same same vibe like being able to lean on like uh

02:35 people that you would never like think of interfacing with like founder of scribd or like

02:41 founder of lever like these are big companies and like very scary companies.

02:46 And they've gone through all these like experiences and they're just there.

02:50 You can just send them a Slack DM or email or just like schedule an in-person

02:53 meeting and be able to just ask them all the questions that you have.

02:57 Like last time I was like, how do I do enterprise sales?

03:00 And Nate was like, here's the breakdown. This is what you do.

03:03 This is what you do. And it was like such an amazing experience just being able to um i don't know just talk to like these really cool

03:10 people like in person and also like so directly and be able to get some of so much like real

03:16 advice from them so i i reckon it has been very beneficial so far and it's only been so little

03:23 time and it's a lifetime thing, right?

03:25 You go through YC, it's not just the bots.

03:27 It is what comes after, too.

03:29 That adds a lot of value.

03:31 That's excellent.

03:31 If there's any in-person component, you went there in person, right?

03:36 Yeah, exactly.

03:37 I'm living in Berkeley right now.

03:39 Phenomenal, phenomenal.

03:41 Good for you, man.

03:41 That's exciting.

03:42 And, of course, along with the open source project,

03:46 there is the company building aspect. And I understand phenomenal. Good for you, man. That's exciting. And of course, you know, along with the open source project, there is the company building aspect.

03:47 And I understand this is new for you.

03:50 What has happened so far?

03:52 How have you experienced it?

03:53 If you could tell us a little bit about that aspect.

03:56 Yeah.

03:57 Yeah, I can't speak too much about the product because we haven't launched yet, but it's

04:01 been really fun.

04:02 Like every day I wake up at, like, 10 a.m.,

04:05 and I grind till, like, 5 a.m. the next day.

04:08 And so it's really cool, because normally, like,

04:11 just coming from as, like, a student perspective,

04:13 where it's like, yeah, maybe I'll go to a class and do the homework

04:17 and then spend the rest of my time, like, coding.

04:19 It's so, like, it's a very freeing experience,

04:23 like, being able to do something that you've always

04:25 tried to do over some like barrier and finally be able to do it like full time.

04:32 I mean, there's a lot of challenges, right?

04:33 Like, how do you like manage a team?

04:35 How do you talk to enterprises?

04:37 How do you talk to customers?

04:38 How do you think about business?

04:40 How do you think about market?

04:41 So there's like a lot of things that like I have to think about, but never had more fun in my life so that's a good thing i think super exciting absolutely

04:51 and uh the the team talking about the team and now with the new funding of course and the new support

04:56 you have the ability to grow so how are you approaching this what's your strategy like

05:01 have new people joined so far or planning to extend it more this year

05:05 yeah so we have two people in the javascript ecosystem uh that are working with us one from

05:12 solid js one from like point managers vtest if you use these tools um and then uh in person i have

05:19 two two good friends like one that works on machine learning and then another one works on

05:24 like infrastructure cloud infrastructure that type of thing and so uh a bit of a skills like range of

05:29 skill sets that i personally don't have and also people i enjoy working with and can can see myself

05:36 waking up for the next 10 years working side by side i love it that's that's that's great and

05:43 going back to what you said about you sales, that whole motion, anything that you've learned so far or something that you might like to highlight from the resources and advice you got from the peers at YC or Gitron?

06:08 things um like being like i guess the first point is like um like one of the most important things is understanding how enterprises think like who are you talking to like why like what are their

06:14 incentives and like what are they concerned with you like i don't know if you like you probably

06:18 probably read like if you're a listener like you probably like uh already kind of know this but

06:23 like enterprises don't want to bet on a startup that can potentially flake in like a year or like die in a year right

06:29 and so like how do you convince them of this and like commercial open source is a great option

06:33 because like you could just say like we'll provide enterprise support as long as we live

06:36 and if we die you're open source anyway so just use us right um but like yeah like longevity is

06:43 like a concern for a lot of enterprises

06:45 like especially in big organizations like their incentives are not the product is good like you

06:52 have to like convince them of another angle for example like oh um excuse me um like

07:00 this will make your developers more productive if you're talking to like a pm

07:05 or something whereas a developer it'd be like this make your workflow better or if you're talking to

07:10 like a head of sales it's probably more like we will make your sales increase by this many this

07:15 much or something like that and so playing to like people's incentives is like huge importance

07:19 for enterprise sales um i also found like for early stage uh it's not it's more about selling yourself

07:27 as a person than selling the actual product like from what i observed like the product is like

07:33 because the product won't be amazing at the time so it's like very hard to pitch a good demo you

07:37 can pitch a demo but like it's not hard to pitch like a oh this is a comprehensive product you

07:41 probably will buy it's more like you know what I want to be here to fix your performance issues.

07:46 And that's kind of like the strategy I'm going with.

07:50 What else?

07:51 I guess like top-down enterprise sales is pretty hard.

07:56 Like especially nowadays, especially in this economy.

08:03 So I found that bottom up is usually easier but

08:06 there's trade offs between those two

08:08 I guess

08:09 that's excellent advice actually

08:13 and you know sounds like it's coming

08:15 from someone who spent 10 years doing this

08:17 yes

08:17 so hats off to you for that

08:21 that's great and then

08:23 all these conversations that you've been having so far

08:24 how did they come about?

08:26 Were they already users

08:28 of the open source software or came

08:30 to you or did you actually do any outbound

08:32 as well?

08:34 I actually did zero outbound for

08:36 this. A lot of it is from

08:38 my personal brand, so Twitter,

08:40 whatever, and then

08:41 GitHub repo users.

08:46 It's really cool that open sourcing do this like i literally just like post like a random tweet like oh i'm working on something

08:51 and people just email me of whatever they're working on so um hats off to open source i guess

08:57 phenomenal and so that brings us to my follow-up question which is about marketing, about building in public, creating content, generating that inbound.

09:07 First thing is, if you have any advice on that,

09:10 because a lot of people are trying to get started doing more of this,

09:12 feel like they don't do enough.

09:14 And then the second thing is, if you would go as far as saying

09:17 that it's actually today a necessity for someone building a startup

09:22 or for an engineer building product to build those

09:25 skills cultivate them and distribute their product to the world this way yeah marketing is a hard

09:33 thing like honestly i am also pretty lucky in this regard like i don't i have no clue how i'm here

09:39 um but uh i think it's um it comes down to a couple things um one is like be very like

09:49 decently unconventional like a lot of people like companies are going for like the linear

09:54 clone approach like in terms of their branding it's like oh yeah we look like linear and it's

09:59 like okay great but also i can't differentiate you from the other 10,000 websites, right? I say this when our website looks like linear.

10:07 But, like, I don't know.

10:11 Differentiating is pretty important.

10:15 Like, I think one of the most important things

10:17 is, like, being able to, like, there's strategies.

10:21 Like, you can, like, for example, with Million,

10:22 what we do is like i post like flashy

10:25 apple videos sort of apple style videos that people like watch and it just gains their attention

10:29 and like wow i'm gonna check this out um uh also like technical blogs are pretty pretty good uh

10:36 that that works for us um but i don't think it's like very universal like one thing is super

10:41 universal though is like branding like like for example like when you think a million

10:45 like you maybe think of like speed fast

10:48 virtual dom whatever but you also think of like

10:49 this person like oh yeah

10:52 this is the same thing right

10:53 and so like being able to like

10:55 have something that's something different associated

10:58 with your product is pretty important

11:00 like I can think of a couple

11:01 of examples like

11:03 like Vercel is like black or white triangle or something other examples like uh like vercell is like black white triangle

11:06 or something like that and then like or uh agolia is like oh the the prettier uh prettier like

11:13 bounty open source thingy or something like that yeah no uh absolutely that that makes sense and

11:19 when it comes to like the actual nitty-gritty of like making the time to post every day or maybe

11:23 producing a super high quality video uh and and when people start off like it feels like you're chatting out in the

11:30 void right and you don't immediately see that return so what would you tell folks uh who are

11:35 at that stage about doing it how much trust they should put into that process and you know there's

11:40 an opportunity cost they don't do outbound don Don't do customer success. They're just shutting out in the void on Twitter or whatever.

11:46 So, yeah.

11:47 What's to say here?

11:48 I think outbound is pretty hard unless you're, like, really, really good at it already.

11:53 Like, I know, like, the mistake a lot of founders do is –

11:56 actually, I don't have enough experience to say.

11:58 But, like, the reason I didn't do outbound was because, like, I did it once,

12:04 and it was, like, super unsuccessful. Like, maybe I got, like, one reply to, like, 100 emails once and it was like super unsuccessful like maybe i got like

12:06 one reply to like 100 emails and i got my email ban so it was not very worth it um like what i

12:15 found was uh people will come to you if they find what you produce is interesting. And that stuff is only interesting

12:26 if it's interesting to yourself.

12:28 For me, it was virtual dom.

12:30 This thing is a super interesting thing

12:32 and nobody knows about this.

12:33 And I was like, here's what I made.

12:35 And people are attracted to that.

12:37 Something that you find generally interesting,

12:41 there are other people who also

12:42 potentially could find it interesting.

12:46 100%. Take us back to like the very first days when you were building the very first versions making the

12:51 very first announcements what did that look like and how much did things change sort of like from

12:57 stage to stage quarter to quarter like during those first years yeah first year and a half basically no stars oh there were stars like

13:07 basically no traction at all uh i was like posting a lot on like dev too if you know that

13:13 also hacker news reddit uh and i was probably getting banned shadow banned on all those

13:18 services but uh yeah just like like i don't know um it was pretty hard to like uh partially because the product the

13:28 actual source product was pretty bad like it wasn't like i advertised it as like fast virtual

13:33 dom but it wasn't really that fast it was actually slower than react at the time

13:36 so like partially that and also uh i was like like i i was like trying at the time, I was just trying to like get traction.

13:48 I was like forcibly trying to get traction.

13:50 And at some point, I was like, you know what, I'm not I don't want to try this anymore.

13:53 Like it sucks.

13:54 So I took a month break.

13:55 I went back to and just started tinkering on like actually deep diving into virtual DOM and just tinkering on that.

14:03 And what I found was that there was this really cool

14:05 project called block dom and block them was a really cool approach to virtual dom that was like

14:10 way faster and today it's like one of like top five fastest frameworks on the benchmark

14:15 um and i was like okay what if i clone block dom but applied it to react and so i spent a couple

14:21 months working on that i released it to the world and it just blew up

14:25 because like people are really resonating with that.

14:27 Like who doesn't want a faster React?

14:29 And so the key insight there was like,

14:31 I could post how much, however much I want,

14:35 but what really attracted people was that key insight of like,

14:38 they can use their existing React, have faster virtual DOM.

14:43 And I, on the side, I i had little tricks like posting technical articles

14:46 or like flashy videos or whatever absolutely so it really starts and ends with a product

14:53 it was the way you said then the rest is auxiliary activities to drive more awareness drive more

14:58 traction and now this is a time where you're just building it for yourself to improve your own skills and to create your network.

15:07 At what point did it occur to you at the back of your head that, hey, there could be something commercial here, an occupation for me, a career to be built here, a company to be created that was recently created?

15:21 So when did that kick in?

15:28 that was recently created so when did that kick in yeah uh frankly like million open source like the open source project is not commercializable i don't plan to commercialize it and i think

15:32 it's better as the public good right like who does i mean like it'd be great like everyone

15:38 has slow react apps like why don't um why don't we like have something for that? So forever remain open source.

15:46 But I was, like, while I was building Million,

15:49 like, while I had traction and people using it,

15:51 like, people started reaching out to me.

15:53 And, like, it wasn't, like, random developers.

15:56 Like, there were a lot of random developers,

15:57 but also, like, some engineers at some very big companies

16:01 started reaching out.

16:01 They were like, yeah, we're trying to use Million.

16:10 And because the React app is really, really slow. was like hmm why is your app slow tell me more um so i kind of went through this process and figured out like there's this huge epidemic

16:15 of like really slow web apps and these engineers can't fix them because of two reasons like the

16:21 first reason is it's really hard to fix performance issues. It takes months.

16:25 People buy Sentry or some random APM

16:29 and just throw it in there

16:30 and still can't figure it out.

16:31 Hire a performance consultant.

16:33 It takes months for them to solve.

16:35 The second part is people don't understand

16:36 how to write performant web apps.

16:38 And honestly, it's not their fault.

16:41 If it was their fault,

16:42 then I probably wouldn't be commercialized.

16:45 I was probably just yelling at them, like some people do.

16:48 But, like, I was like, okay, this is an issue.

16:51 People, like, really, really, really want to solve it.

16:54 And, you know, web performance directly leads to higher sales, better user experience, better user retention.

17:00 So there's, like, real for higher, faster web performance.

17:07 So I was like, okay, you know what?

17:09 Let's try to build something out of this.

17:12 And of course, you're not going to help

17:14 but ask what forms this can be taking.

17:18 First off, like my understanding is that

17:20 when all these companies,

17:22 all these engineers from these big companies came to you,

17:24 some sort of like consulting arrangement could come about

17:27 where, you know, you're actually digging in, helping them improve

17:32 their code base, improve their performance. And then there could be some

17:35 user-facing tools, you know, with AI, for instance, where you could

17:39 actually improve the performance of your application, more hands-off approach.

17:44 So these and any other sort of like formats

17:47 that came on the table, thinking about exploring,

17:51 what could you share today about this?

17:54 We're currently going with a top-ish down approach,

17:58 just a little bit of a mix of like middle to top

18:01 versus top-bottom,

18:03 mainly because large enterprises make money. That's just the

18:07 reality. And also large enterprises have the highest need for web performance. So it just

18:13 like happens to be a great market for us to go into. One thing is like large enterprises have

18:22 high complexity code bases. So like, you know, like normal startups or small projects,

18:26 they're pretty clean.

18:27 There's Next.js project or React.

18:29 Like some of these large enterprises

18:31 have extremely complex pipelines.

18:33 Like they use things to render other things

18:35 that like you've never think should happen

18:38 or actually it probably shouldn't happen,

18:39 but they're doing it anyway.

18:41 Like they're either like an old React version

18:43 or they're using some backend frameworks

18:45 to render frontend stuff.

18:46 And it's a huge technical debt.

18:51 And so being able to solve that for them

18:54 is extremely, extremely valuable.

18:58 That's how we're thinking of it right now.

19:02 Absolutely.

19:03 And I mean, so it is a work in progress when it comes to like the formats this will take. And of course, you know, you said earlier that Million, you want it to be a public good, something everyone has access to for free. But of course, this gives you the opportunity to build a lot of auxiliary services and new products around it, having the food in the door and replacing people's code base and minds.

19:26 And so, I mean, that's something to bet on.

19:29 And it makes total sense that you went down this commercial path

19:33 and why you have these investors backing you.

19:36 What else has been new for you over these past months

19:40 when it comes to your daily activities or something?

19:44 Were you're like oh

19:45 five months ago or eight months ago i was totally clueless about this i wasn't more on oh wow there's

19:50 a new world here i didn't understand uh if you could share something like that uh probably a lot

19:57 but also very little like i probably learn like new things every day. It's like kind of crazy. I don't know.

20:05 Like if you're skipping school to like work on your project or product,

20:13 like probably should like gap if you can.

20:16 Like especially if you have, I don't know, like during school,

20:20 I was pretty like risk adverse.

20:22 I was like, I was working on this thing.

20:23 I was skipping lectures, but I wouldn't like accept the fact that I should probably should gap or drop out.

20:30 And like, I had all of these cards in my hand where I was like, you know what? Worst case

20:34 scenario, I just get a job or just return to school. I don't know. Like if you're,

20:39 if you're someone in that position, you probably should seriously consider it.

20:42 At least for me, it was very valuable.

20:44 position you probably should seriously consider it uh at least for me it was very valuable um

20:55 uh reading is good i didn't read before at all uh and i had like i met a friend who like like sends me random books and articles i've been starting reading and like i've learned a lot from

21:00 that um i don't know talking to customers is good as well.

21:05 I used to build things that are magical the first year and a half.

21:08 I just built things that are based on an imaginary thought in my head

21:12 that I wouldn't do it for once.

21:13 And after that, I just realized what they actually need.

21:19 I think the last thing is...

21:21 I was just talking about this, actually.

21:27 Generally, people... Well, well actually i don't know like if you're a person who makes like very rational decisions like you should

21:32 you should stop making like you should open your bounds of rationality so like normally people like

21:38 they close their bounds to a certain like okay based on these conditions i'm going to make this

21:42 rational decision i found like a lot of the decisions I made in the past six months

21:47 were extremely irrational.

21:49 Applying to YC last minute or applying to GetRound

21:53 or dropping out of school were pretty irrational decisions,

21:57 but it just happened to work out.

21:59 And I guess sometimes you need a little bit of randomness

22:03 to get you further.

22:06 That's two words of an entrepreneur, absolutely.

22:09 And I feel like sometimes we even self-sabotage with, you know,

22:13 personalities who talk ourselves out of doing, you know, what needs to be done.

22:18 And I feel like, especially if you're starting a startup

22:20 and you're like a very strong engineer who could be, you know,

22:24 working at a very big company, could be you know working at a very

22:25 big company earning you know very big salary and so on and you're like oh now i have to do this

22:30 shit work which is like because a lot of times a lot of times like something yeah the job is a lot

22:37 of shit work right um but but i feel like yeah if we just silence the like rational part being like

22:43 what the hell are you doing here like this doesn't make any sense and we just silence the like rational part being like what the hell are you doing here

22:45 like this doesn't make any sense and we just grind it through we can be surprised with what

22:50 comes on the other end and it does take time i mean how old were you when you wrote your first

22:54 line of code probably sixth grade it was sixth grade on my like libraries dell desktop wow did the did your family like sort of like have

23:07 a background there and you followed or is it just something you picked up on your own and

23:12 found your interest earlier yeah i grew up with a computer so like the first time i actually

23:17 touched a computer was when i probably touched a computer before that but like like um like first

23:22 time i actually used computer was like my school's libraries like desktop computer um that but like like um like first time i actually used computer was like my school's

23:25 libraries like desktop computer um i remember like being in like not even notepad plus plus

23:32 like notepad windows notepad and writing my first html file and it was really fun really fun

23:37 did you have any classmates or peers where you were like geeking out on these things together

23:42 or has it been more of a lone wolf kind of journey for you well it's a lone wolf yeah it's like younger like i really

23:51 struggled to find like community like people were interested like people around me like

23:55 i don't know like they had xboxes or play games or whatever i was like you know what i like i like

24:00 doing this weird thing where um i i make put these put these weird like HTML thingies and then it just outputs cool websites.

24:10 Yeah.

24:13 And to people who are, it's crazy to say this,

24:17 but who are like still in middle school or high school or people like in

24:20 college. And I really hope that, you know,

24:23 when we publish this interview,

24:24 this can act as

24:25 an example to them of an alternative way right um what would you say to them directly and you

24:31 don't have to tell them oh drop out of school and like go do this what are the risks involved

24:36 or are there really risks involved because you know like when I started as an entrepreneur like

24:43 as you said similarly like worst case scenario I get a job in the end.

24:47 And even if this fails, it feels like I will come out sort of like better and stronger as an individual, the skills I will build, the networks I will build.

24:56 So, yeah, what would you share to people kind of like around your age who are contemplating startup, career in software or building their own things

25:05 yeah depends on what your incentives are i remember jay telling me this actually uh

25:10 joseph jacks os's capital he's like we had this one hour meeting and i was like so tired but like

25:17 i remember him telling me like yeah aiden you know my you know my threshold for knowing when

25:22 someone's like founder or not whether they

25:25 can't like whether they can't stop thinking about it at all times like like um if you're in like

25:32 i don't know like what would i what was the shit i pulled like if you're like in the hallway like

25:39 hiding in a corner like coding on a million or your own projects like that's probably a good

25:44 sign you're gonna be a good founder.

25:46 But if you're just

25:47 LARPing for, I don't know,

25:50 LinkedIn likes or something like that, it's probably not a good idea.

25:54 Totally.