My research interest is systems for machine learning. I look for scalable and programmable abstractions that bind accelerators, models, and application logic together. I believe that a good system empowers its users, embraces heterogeneity, and adapts to change—all without fighting physics.
Recently, I have been applying these principles to
Pie, a programmable LLM serving system that enables dynamic application logic to run natively within the inference engine, unlocking new possibilities for AI agents.
My broader interests span programming languages, operating systems, neurosymbolic computing, and game engines. At Yale, I am fortunate to be advised by Prof. Lin Zhong.
News
Publications
Education
Yale University
Yonsei University