I'm a founding member of J.P. Morgan AI Research, where I lead a team building AI systems that learn how financial markets work – from modeling how individual trades form prices to optimizing how portfolios are built.
Most recently, we built TradeFM, a foundation model that learns how prices form from billions of trades. We also work on reinforcement learning for portfolio optimization and knowledge graphs for financial analysis.
Originally from New Delhi, I studied Computer Science at Georgia Tech (M.Sc., B.Sc. Highest Honors), where I researched reinforcement learning across domains – language model alignment with Mark Riedl, equities portfolios with Tucker Balch, and domain adaptation with Charles Isbell.
Outside of work: scuba diving, a growing (astro)photography habit, cortados, and an unresolved Old School RuneScape problem solution.