I'm 20 years old. I've built 3 startups. Nobody uses any of them. Here's my honest story. I will not promote
1. UniConfess
Imagine Twitter but only for college students. Post publicly or go fully anonymous — your identity is never revealed. You get two feeds: one just for your campus, one global so your voice reaches students across every college in India. There are clubs, direct messaging, a marketplace to sell old books and electronics within your campus, leaderboards, and a full admin panel where I can assign sub-admins per campus to moderate content.
I built the entire thing solo. 14,500+ lines of code. Hybrid React + Flutter architecture. Supabase backend. Firebase push notifications. It's live on the Google Play Store right now. It covers 71,000+ colleges and universities across India.
Downloads? 25. All my friends.
The app works. The product is real. It's just sitting in some corner of the Play Store that nobody visits because I have zero idea how to market anything.
2. Scaling Yug
I started scraping Hinglish data — Hindi + English mixed, the way a billion Indians actually talk. Got to around 4-5 million rows. Then hit a wall: cleaning and labeling millions of rows manually is impossible.
So I built a pipeline. An automated one. It cleans, deduplicates and labels every row without human intervention — assigns intent, emotion, toxicity, sarcasm, quality score, label confidence. All automatically. The pipeline runs at 100,000+ rows per second.
I now have 1.5 million rows of cleaned, labeled Hinglish data sitting on my laptop.
Market value? Around $25,000–$50,000 for non-exclusive licensing. Over $100,000 exclusive. There are AI companies literally searching for this kind of data — Indic language models are starving for it.
I sent cold emails. LinkedIn DMs. Got zero responses. So I gave up and closed the laptop.
That dataset is still there. Untouched. Worth potentially six figures. On my laptop.
3. MediYug
This one hurts the most because the problem is genuinely serious.
People are going blind. Going untreated. Not because medicine doesn't exist — but because they don't know where to go or can't afford local healthcare. Medical tourism is real and it's growing, but there's no good resource for normal people to navigate it.
So I built one. MediYug helps you find affordable hospitals in other countries based on your medical condition. It compares costs, gives you visa information, travel tips, expert advice. Everything in one place. No more Googling 20 different things and still being lost.
I finished it. It actually works.
I launched it without a domain on vercel for free because I can't afford a domain(My father is a rickshaw driver and over monthly income is around $100-$130)and $90 is emi. Just threw it out there to help people. Nobody found it. Nobody used it. Because I told nobody.
So here I am. 20 years old. Three real products. Three real problems being solved. A dataset worth potentially $100k+. And the combined user base of all three is roughly my friend group.
I can build anything. I genuinely cannot tell anyone about it.
I'm not posting this for sympathy. I'm posting this because I think a lot of builders are in this exact position and nobody talks about it honestly. We celebrate the launches. Nobody talks about the products that work perfectly and sit completely invisible.
If you've been here — how did you get out of it? What actually worked for you?
And if anyone in AI/NLP knows what to do with a 1.5M row labeled Hinglish dataset — I'm all ears. It deserves better than my laptop's C drive.
