The Brief
HumanAIze Hackathon (FinTech Edition), hosted on Hack2Skill, gave teams a short window and an open problem space in fintech. My friend @Yameen and I picked something ambitious for a hackathon timeline: an AI-powered stock market analyst, focused specifically on the Indian market — giving stock suggestions, specific predictions, and real-time market news in one place.
We called it Bronn.
The Split
This was a two-person team, and the split was clean:
- Yameen — backend
- Me — frontend (React + React Native), plus most of the product work: ideation, discovery, technical feasibility, and research
The product phase mattered more than usual here because of the format. In a hackathon, you don't get to discover halfway through that your idea doesn't hold up — you have to stress-test the concept fast, figure out what's actually buildable in the time given, and commit. That's where most of my hours went before a line of frontend code got written.
What Was Under the Hood
- Frontend — React for web, React Native for mobile, so the experience worked across both from day one
- Market data — Yahoo Finance API for live pricing and market movement
- Analysis/predictions — Llama 3, run through Groq's API for fast inference, powering stock suggestions and predictions in real time
Wiring real-time financial data into an AI-driven suggestion layer, inside a hackathon clock, was the core technical bet — and it worked.
The Outcome
Bronn didn't advance to the final round. But the feedback that came back from judges was that we were close — the project was well received, and we finished in the top 90 participants, a strong showing that told us the miss was narrow, not wide.
That distinction matters. A hackathon project getting complimented by judges and falling just short tells you something different than a project that simply didn't land. It told us the idea and execution were sound — the constraints of the format were the limiting factor, not the concept.
What I'd Point To
Bronn is a good marker of how I work under pressure and ambiguity — same as Value Islam, but compressed into 48 hours instead of a full project timeline. Owning the product phase (feasibility, research, scoping) while also shipping the frontend across two platforms, in a hackathon window, with a partner handling backend in parallel — that's the case study here.