CS undergrad at Columbia (Egleston Scholar, top 1% of class, GPA 3.8). IEEE-published in deep learning, USACO Platinum perfect score, and currently a Software Development Engineer intern at Amazon. Fluent in Python and C++; love tough puzzles.
I work on machine learning, quantitative finance, and the systems that hold them up. The problems I like are the ones where the math, the model, and the infra all have to agree.
At Amazon this summer I’m on the Long-Term Planning and Forecasting team in Supply Chain Optimization Technologies (SCOT), building the centralized Config package that all 13 engineers use to drive MOSAIC, our long-term revenue and inventory forecasting platform.
Alongside that I build Quantiv, an options-implied earnings platform on Vercel: multi-week calendar, screener, symbol pages, and a Clerk watchlist over nightly JSON, with LightGBM scoring on DuckDB / Parquet and live quotes through Upstash. Before Columbia I published single-author at the IEEE ITSC on graph-network traffic forecasting (24% RMSE win over STGCN) and built a CFD + neural-net F1 wing optimizer.
Apps in production, papers that have been published internationally, and competitions I'm still proud of. Look here to see what I built, what I used, and what I learned.
Building the shared, version-controlled configuration layer behind Amazon SCOT's long-term revenue and inventory forecasts, replacing scattered service setups with one reliable source of truth.
Peer-reviewed and competition-published research.
Open to internships, collaboration, or a quick coffee chat.
