I just wanted to document my journey over the past 6 months or so, because it was real. And I wanted to put out whatever I learnt over this journey. Or the best case, maybe it will inspire someone to take action on that one idea sitting inside their head for a while.
How it all started
I picked up coding in February 2025. Vibe coding was just blowing up, Cursor had launched and was giving out a ton of free credits, and I just downloaded it one day and built my entire personal website in three days. I didn't understand a single file that was being generated. But something clicked. I realised I was built for building things and putting them out into the world.
So I started doing exactly that. Side projects, Twitter posts, whatever I could ship. But by around June 2025, I wasn't really satisfied with what I was doing. I felt like I was operating way below the level I was capable of, and I wanted to do something much much more.
Then came the flash of inspiration.
I remembered my JEE prep days. There was this math professor who would answer my doubts over WhatsApp at any hour. Just send him a picture of the problem and he'd explain it. Not everyone has access to that kind of teacher. I also remembered a platform called Doubtnut, where online tutors would personally record and upload explanation videos for student questions. And I thought why not automate this entire thing with AI?
That was the seed. What if there were AI tutors that could explain any concept or question you typed, with a full animation like 3Blue1Brown, but on demand for anything?
That was zTutor.
Building the first version
The first version was embarrassingly simple. You'd type something, and you'd get a 10–20 second animation. Just some pretty animations.
But I still remember the moment the first animation rendered after hours of debugging. It felt like magic. I hadn't written a single line of code myself, but something beautiful came out on the screen.
From there it was gradual. First video, then audio, then the hardest part: syncing the two. You don't want the visuals and narration to mismatch while a teacher is explaining. I tried around 10 different architectures before I cracked it. Eventually, the product could generate 1–2 minute fully animated, narrated explanation videos for any topic you entered.
Launching it
I made one early mistake. I spent too much time building and not enough time putting things out in public. Eventually I realised I had to show it to the world.
There was this friend of mine, Tushar, who used to run these demo day sessions for founders. I recorded a demo video with him. I was really hesitant. I had never really put myself on camera before. But I pushed through, launched it on Twitter, and the reactions were amazing.
Then came the waitlist, then the full product launch in December 2025. I made a launch video, edited it properly, and posted it everywhere: Twitter, LinkedIn, everywhere. I was a pretty introverted guy back then, but at that moment I didn't feel any hesitation at all. I just put it out. And people showed up.
Getting backed and the first pivot
zTutor started with students as the primary market. We had pricing, we had users, we had some early revenue. But I quickly realised students weren't willing to pay much for just a video generator, especially when bigger platforms were offering study notes, flashcards, podcasts, and videos all bundled together.
Around this time, I applied for a grant from 3F Venture Capital, a micro grant for young student builders. After five or six calls, I got it.
This was a big moment for me. For the first time, someone believed in my idea enough to put money behind it. That person was Yash. Yash, if you're reading this, thank you for believing in me early.
Yash suggested I pivot to B2B. Sell video as a service to other businesses instead of directly to students. So I looked at the data analytics space. There were 10-20 companies building tools where you could query your database or knowledge sources and get dashboards. I emailed every single CTO and CEO I could find. I couldn't close any of them. The market wasn't right.
The SAT pivot and the near-misses
Next pivot: SAT and JEE prep markets.
I emailed every major edtech company in India for the JEE market. No actionable response. Then I went after every SAT company in the US, from small startups to big players. I didn't leave a single one out.
This is where things got interesting.
The CEO of Magoosh was genuinely impressed. We got close, really close, to integrating zTutor internally into their platform. It fell through because AI integration wasn't on their roadmap. But the CEO believed in what I was building, and that meant something.
I also sent a demo to the Head of Education at Anthropic, the former co-founder of Schoolhouse. He said it was an amazing product. Moments like that made everything feel worth it. An idea that started in my head was now being validated by people who had accomplished so much more than me.
And then I closed a paid pilot with a YC W25 company. A real, paid pilot. That was a win.
The decision to stop
But I couldn't convince other clients after the pilot. And I started noticing something that I couldn't ignore.
The moat was fading. Video generation, which was zTutor's core differentiator, was becoming less and less of a unique advantage. Models were getting stronger. Notebook LM from Google was doing more than what I was doing, for free. Students and businesses wouldn't pay for something they could get better elsewhere at no cost.
I kept pushing for another month or two. But eventually I had to accept the reality: it was time to stop working on zTutor.
This was painful. More than I expected.
My entire identity had been tied to zTutor. People recognised me by something I built. It had been six months of everything: late nights, cold emails, technical failures, near-wins. And I had to let all of that go.
But I knew it was the right call. And I made it.
What came next
I wanted to take everything I'd learned and apply it somewhere I could grow fast. I got an internship at Rowboat Labs, a YC S24 company, working directly under the CEO and the CTO. I'm also running Eunice Labs, my own independent AI research lab.
The journey continues.
If you're reading this
If you have an idea sitting in your head right now, just put it out. You don't know what it's capable of until real people start using it.
There will be hesitation. A lot of it. Don't listen to it.
And if you have a strong enough belief, you can push through more than you think.
A lot to build in the upcoming times. Thank you for reading!
Gagan