Hi, I'm Ved, a second-year Computer Science PhD student in the Theory and AI groups at Columbia University, where I am fortunate to be advised by Adam Block. I am also affiliated with Columbia ARNI, and work closely with Max Dabagia.

My research focuses on reinforcement learning and its applications to language models. I’m particularly interested in using theory and empirics in tandem to understand why algorithms work in practice and how they can be improved. I’m currently working on the theoretical underpinnings of on-policy distillation, disentangling the factors that make this popular post-training technique so effective.

I received my B.S. in Computer Science from Cornell University, where I worked with Karthik Sridharan and Sainyam Galhotra on theoretical machine learning.

Feel free to reach out at ved@cs.columbia.edu.