~$
Based in Berkeley, CA. I split my time between building, founding, and researching. I write occasionally.
I'm a CS student at UC Berkeley, currently doing ML research at the University of Louisville on explainable AI, focusing on why vision models confuse similar-looking things, like a 3 and an 8.
I started coding in middle school, trying to rebuild Chrome's dinosaur game in JavaScript. It didn't work well, but it was enough to get me hooked.
Years later I noticed schools don't teach financial literacy, so I built ProfitQuest to teach it in classrooms directly, which won the Congressional App Challenge. That led to MoneyBot, a separate, funded startup doing the same at a bigger scale, where I worked as a founding engineer.
If you're building something interesting, reach me at suriyansn5@gmail.com.
work
ML ResearcherUniversity of Louisville
2025 — present
Founding EngineerMoneyBot
2025 — 2026
activity
projects
ProfitQuest
A web app teaching high schoolers to budget, invest, and think about money before life makes them.
Spiral & Wave
Catching early Parkinson's in the shake of a hand-drawn spiral.
Alloc
Learns how to split a portfolio, then actually sticks to it.
Filter
Reading bloodwork to catch kidney disease before symptoms do.
view all projects →
writing
Why models confuse things that look alikeJune 2026The subject schools forgot to teachMarch 2026Notes on mechanistic interpretabilityNovember 2025
press