Twincloud's Youtube Subtitle Extractor
12:46
A few weeks ago, I got a DM on X from this guy named Yasser. We got to
chatting and he told me something I couldn't believe. He took his app from 0
to 1 million in just7 days. And what's even crazier is he did
it with zero audience. When I launched, I only had 16 followers. I brought Yaser
onto the channel to talk about how he did it and he shared everything,
including a breakdown of the tweet that changed his life, how to make millions
with AI agents, and what most people get wrong when it comes to building their
MVP. This one you can't miss. I'm Pat Walls, and this is Starter Story.
All right, Yaser, welcome to the channel. Tell me about who you are, what
you built, and what's your story. Okay, so my name is Yaser. I built Chadbase to
7 million in ARR with no funding. I launched it 2 and a half years ago as a
soloreneur to my 16 followers on Twitter and it went viral. Wow. Okay, that's
insane. I haven't seen numbers like that before. Tell me about how you came up
with this idea for Chatbase. So I followed the bootstrapping movement on X
like Peter Levels, all those guys. I was doing actually internships in fang
companies but once I got there I realized maybe it's not for me. I looked
at people five 10 years ahead of me in their career and it was not something
that I was looking up to but when I saw like people like Peter, people like
Danny Pusma especially in the AI wave I thought hey like I can actually do this.
I saw people doing demos of how you can build a chat GBT for your data and stuff
like that but no one was building a product around it. I knew like this is a
good idea. I didn't do any validation or anything. I just like said if I don't do
this someone else will. So I released chatbase to my 16 followers. It went
viral. All of this was happening in my final year of university, this was just
like a side project, right? I was happy it made money, but I'm not going to drop
out yet. But when I saw the first sale and then like it kept going and like
this is this is significant. I had like two or three sales every day and uh I
dropped out. All right. So, you mentioned that you tweeted something out
to your 16 followers. It went viral and essentially that this this one tweet
kind of changed your life. Can you walk me through this viral tweet? I think
this was like my third tweet ever in my life. For this one, three things made it
work well. One, a familiar interface, the streaming interface where like the
text is streaming uh in a in like a chat interface. Yeah, this was like a very
chat GPT thing which like at the time was going viral. So, I think taking a
glance at the demo, it's just just interesting. The other part was
minimizing the seconds to like the aha moment. I think I did that well here.
Like it's a very simple demo, right? But it just shows you exactly what I built
was capable of in like the first 20 seconds, right? And I think the other
thing I did is mention the tools that I was using 2 and a half years ago. Like
there was a bunch of new AI companies, AI tools starting out. No one had any
customers because they're all new. So every company was trying to show people
like, hey, this is something that is being used by real people building like
real products. When I mentioned like Langchin and Pine Cone and stuff like
that, it made sense for them to push it too because no one else was building
stuff like this on their platforms yet because it was all very new and they
they wanted to promote like their AI features and stuff like that. When I
mentioned this, those people picked it up and they were like going viral. So my
thing also went viral and then it created this viral loop and I think yeah
most of most of the initial traction was from that. I really do like the kind of
build in public thing that you're doing. You're kind of sharing what tools you
were using and that was a big piece in it. if you could break down the build in
public playbook because I know you kind of did this over and over again. Tell me
what you learned and and what works there. Yeah, so I think the easiest
thing is just to share what you're doing every day. That's like the most basic
thing you can do. First two or 3 months if you're if you're not lucky, it will
just be you practicing what works and what doesn't. It's it's not going to get
any any views because you're you're trying to build a brand, you're trying
to build a story, one tweet is not going to do it, right? People need to see your
face a lot. Preferably, if you can do videos, too, post them on X. a few
people in the build and public space are doing that and they're doing very well
because not enough people are doing that so you stand out more. Frame it as like
I'm building something. I'm learning a lot along the way. I'm going to take you
on that journey. I would say also an important part is not to be boring. I
think it's better to be more personal. Share like your personal story. Have a
personality like have controversial takes maybe if if it's something you
actually believe in because that's how you get people to notice you. All right.
Before we get further into how Yasser built this business to over 6 million
ARR, I want to call out something a lot of early founders forget about your
bank. Yeah, I know not the sexiest part of your stack, but when your app goes
viral, revenue jumps, and your user base explodes overnight. The wrong bank can
become a massive bottleneck. Low card limits, slow payouts, manual everything.
That's what happens when you use a finance system that wasn't built for
startups. That's why we've partnered with Brex, the modern finance platform
designed for builders like you. With Brex, you get up to 20x higher card
limits, built-in expense tracking, automated payments, high yield returns,
and sameday liquidity, so you can move fast when the moment hits. One in three
US startups already use Brex, and they put together this free guide for you,
the startup's guide to modern banking. It breaks down how to extend your
runway, protect your cash, and avoid the kind of finance headaches that kill
momentum. Just click the first link in the description to grab the guide. Thank
you to Brex for supporting the channel. Now, let's get back to Yaser's story.
Before we jumped on this call, you told me something crazy, which was that
Chatbase went from zero to a million ARR in 117 days, which is less than half a
year, just a couple months. That doesn't even seem believable to me. So, I'd love
if you broke down how that's even possible, that an app can grow that fast
and and how yours did. After the first initial viral moment, I kept adding more
features and with every new release, I would frame it as a new launch. I think
some people when they tweet, they frame like the second or third launch as like
feature launches. So, if someone new is looking at that, they don't have any
context, right? So, they're not they're very likely to skip it. So I try to
frame every single launch to make it make sense to new people that don't know
me, that don't know Chadbas. The other thing was just like being creative
around going viral without spending any money. So what I did for example going
into subreddits of books, I would create a chat GPT for their book. It was like
for free. I was losing money on this. The only point of doing this is to have
my domain opened by a lot of people so I get like more domain authority. Maybe
they click on it and then they see the landing page. they see, hey, like I can
actually pay money for this. I kept doing that for all the books. I did that
for a lot of influencers, too. I did this for Paul Bram, for Naval. Lastly,
also sponsored posts were very big. If I'm not launching something today, then
I have to do a sponsored post with a big page. We did one with a page on LinkedIn
that day. I think we did $4,000. So, so yeah, I think sponsored posts, if you
structure them well, they're a very good way to get your product out there.
Awesome. All right. Well, can we dive into some of the numbers behind this
business? How big is this thing become? Yeah, right now we're exactly at 6.8
million ARR. It's growing very fast. And right now it's only PLG. We just passed
actually 2 days ago 10,000 customers that are paying us between $40 and $500
a month, which is insane. And I think registered users is around like 600,000
something. Okay, that's insane. You mentioned PLG. What is that? So this is
productled growth. You just have a Stripe button on your website. You do
most of the marketing outside of your website. You just have a good landing
page. People can just go in, buy a subscription, put in their credit card.
You don't have to have a lot of features. Just have like one core
feature. In my case, it would just upload a PDF and talk with it. Just have
like a good UI. Don't have any bugs. Have like a good MVP and then launch it
into the universe. Find the balance between launching fast and not launching
garbage. have a good quality tool so that you're proud of when people are
using it and then people would be excited to use it more if they're not
like frustrated using it. All right. Well, let's talk about AI agents. That's
what you've been building for a long time. You know a lot about this. How can
people watching this learn how to build with AI agents? What's the right kind of
framework for learning these and where to go? If you want to build products,
then you go to sources like OpenAI's cookbooks or like their documentation.
Read about their agents SDK. do the same with Verscell has a library called the
AI SDK which is very powerful. Uh just read like their documentation, look at
the examples, look who's using them. Learn how those companies tell you how
to build AI agents. I think actually maybe a good idea is to find something
that you can't build yet and the only bottleneck is the AI model itself. like
the intelligence is not there yet and then start building that and just like
bet that in a year you're going to have GPT5 which is capable of doing that and
then you're going to have one full year like a head start on everyone else. I
think something very similar happened to me. I started working on chatbase before
that chat GPT API. So I had like the product ready. I was people were using
it and then all of a sudden chat GPT launched their chat API. So I just
released a tweet that said in the last 48 hours I I built like this cool new
product and because everyone was talking about the new chat GPT API and how what
you can build with it. This also went very viral. So yeah, I think I think you
can do something similar. All right, let's talk about tech stack. Uh how did
you build this app? What kind of tools and languages and stack did you use to
build this? Yeah, so it changed over time. Now I think the stack we landed on
is Versel for hosting. We're also using their AI SDK for a lot of our AI
infrastructure. We're using superbase for the database. And then we're using
all the AI labs. So, OpenAI, Anthropic, Anthropic, Coher, Google Gemini, Gro,
we're using all of them. So, in chatbas, you can have a drop down and choose
which model works best for you. And of course, Stripe for billing, Dub for
affiliates, Mentlify for our documentation. Cool. You have built this
amazing business. You've been doing it for many years. You've had some
successful things, some not so successful projects. What have you
learned through the process that surprised you? Yeah. So, I think going
into it where we assume that there's like a blueprint on how to build a good
business. It's just who's going to be working hard. But what I realized is
that there's so many different ways to build a good company from like how you
work together, performance reviews, in person or remote, what's the culture
like, all of that can change a lot between company to company. This is
actually good because this means that you don't need a lot of experience to
start a company. Mostly the advice you're getting from people that have
done this before. It's it's just good to listen to but mostly irrelevant because
no one knows your situation as much as you. The best practices change so much
from situation to situation. The people that give good advice will tell you the
same thing. They would give you what they think is right, but in the end they
will say actually like you should like trust your gut mostly because you know
your situation more than me. Cool. I love that. Last question that we ask
everyone who comes on Starter Story, if you could go back 2 and 1/2 years or
however long it's been, stand on Yaser's shoulder, what would be your advice to
young Yaser? Yeah, I think for me specifically, I wasn't thinking big
enough in the beginning. My goal was to just get my 10K, move to Bali, live the
indie hacker lifestyle. I just realized like why not shoot for the stars, right?
Why not build the first 100 million ARR bootstrap company. I was too worried
that because everyone was saying it's like a chat GPT rapper. It's not going
to work. This this got into my head because it was true. It was just a chat
GPT rapper. So, I think I wasn't aggressive enough. I was too shy about
what I'm building. I was too reserved. I think if I go back, I would have moved a
lot faster. I would have put more on the line. So, yeah, just aim higher, I would
say, for me. Cool. Well, that's great advice. Thank you, Yaser, for coming on
to Starter Story. It's amazing what you built. One of the probably greatest
success stories in the India hacker world. and it's going to keep growing
and I love it. So, thanks for coming on. Thank you for having me. Yaser is proof
that you don't need a big audience to get started. He started his app when he
had 16 followers on Twitter and in less than 3 years he grew it to over 6
million ARR. He got started just by building stuff. And nowadays, AI has
made it easier than ever to build. That's why I launched Starter Story
Build. It's our flagship program where in 2 weeks you'll find and build a
simple app with AI. Hundreds of people have joined and launched their first app
into the world. Head to the link in the description to learn more. Thank you
guys for watching. I'll see you in the next one. Peace.