I recently defended my Ph.D. in Computer Science at Stanford, where I worked with Prof. Christos Kozyrakis on software systems for generative AI at scale.
My broad research interests include distributed systems and cloud computing β in particular, I am interested in the system-side problems associated with learning, deploying, and operationalizing generative models at scale.
For my most up-to-date CV, please see my website.
- November 2026: π Successfully defended my Ph.D. thesis, "Adapting to Perturbations: Software Systems for Reliable and Efficient Generative AI at Scale"!
- October 2026: π PACE has been accepted to NeurIPS 2026!
- September 2026: π Check out our new work, Purlin: Separating Orchestration from the Datapath of Collectives! Code.


