My research lies at the intersection of Computer Systems and Machine Learning. I design and build software infrastructure that makes large-scale AI models (LLMs) efficient, scalable, and programmable.
Recently, I have been focusing on programmable LLM serving systems to bridge the gap between static inference engines and dynamic application logic.
At Yale, I am fortunate to be advised by Prof. Lin Zhong.
News
Aug 2025
Cacheback accepted to EMNLP 2025 (Main).
Jul 2025
Pie accepted to SOSP 2025.
Apr 2025
Serve Programs, Not Prompts accepted to HotOS 2025.
Feb 2024
Prompt Cache accepted to MLSys 2024.
Publications
Serve Programs, Not Prompts
HotOS 2025
Asynchronous LLM Function Calling
Preprint 2024
Prior to Yale
Memory-Efficient DNN Training on Mobile Devices
MobiSys 2022
Code
Education
Yale University
2022 - Present
Ph.D. in Computer Science
Yale University
2024
M.S. in Computer Science
Yonsei University
2021
B.S. in Integrated Technology
Work Experience
Apple
Summer 2025
AI/ML Research Intern
Teaching
Principles of Computer System Design
Fall 2024
Yale CPSC 429
Intro to Systems Programming
Spring 2024
Yale CPSC 323
Invited Talks
Rethinking LLM Serving From the Application's Perspective
Oct 2025
Seoul National Univ, Yonsei Univ, NAVER Cloud, Furiosa AI
LLM Prompt Caching
Sep 2024
NVIDIA