About

I'm a fifth year PhD student at the University of Pennsylvania advised by Professor Mayur Naik and Professor Eric Wong. I interned on the AWS Fundamental Research Team for the last two summers advised by Matthew Trager and Stefano Soatto, where I worked on LLM uncertainty quantification and experience-guided reasoning. For undergrad, I attended UCLA where I received my BS in CS in 2020. My research is supported by the NSF Graduate Research Fellowship Program.

My research aims to make AI systems behave as intended. I do this by interfacing code with foundation models, creating agents and workflows which we can interpret, control, and verify. My position on the role of symbolic abstractions (code) in the foundation model era is presented in a pre-print, and my follow-up work explores a core challenge in this space, per-instance program synthesis. Recently, I've been exploring how to interpret and control foundation models in lightweight, targeted ways, and how to enable general AI systems (agents, workflows, and pipelines) to learn from experience so that they become more reliable and lower cost over time.

My publications are listed below, and I have highlighted a few key ones most relevant to my current research interests.

Research Summary

The Interface Between Code and Foundation Models (FMs)

🧠 Concepts as symbols

Surface, compose, and manipulate latent & prompt concepts

Pre-print '25 ICML '24 Pre-print '25 ICLR Tiny '23
Show papers

🎛️ Controlling / interpreting FMs with concepts

Steer, align, and localize concepts

Pre-print '25 NeurIPS MechInterp '25 ACL '25 NeurIPS XAIA '23
Show papers

💻 Programming with FMs

Programs as the execution interface; enforce correctness of reasoning

NeurIPS '25 AACL '24 OOPSLA '24 VLDB '24 AAAI '23
Show papers

Recent News

Pre-Prints

Conference Papers

Workshop Papers

Student Mentoring