About Me

Hi there! đź‘‹ Thanks for dropping by.

I am a first-year Ph.D. student in the Sky Computing Lab at UC Berkeley, advised by Professors Joseph Gonzalez and Ion Stoica. My current research focuses on LLM agents, especially their tool-use abilities.

I lead the Berkeley Function-Calling Leaderboard (BFCL) project, which has become the de facto benchmark for function calling. Check out our blog posts (BFCL v1, BFCL v2, BFCL v3) and explore the codebase—it has already earned 12 k+ stars!

Prior to my Ph.D., I completed my undergraduate studies at UC Berkeley, majoring in Computer Science (Honors Program) and Data Science. You can find my coursework here. I have been fortunate to explore a wide range of domains in computer science and conducted research in computer vision, autonomous driving, and adversarial machine learning before discovering my passion for machine-learning systems.

Teaching is another cornerstone of my educational journey. I served as the Head TA (Content & Infrastructure) for CS 61A, Structure and Interpretation of Computer Programs, for seven semesters—an absolutely amazing experience that let me help students learn and grow. I was also on the course staff for Data 100, Principles and Techniques of Data Science, for two semesters.

Outside academia, I have worked in fast-paced SaaS start-ups. As a founding software engineer at Markit.AI, I developed image-saliency detection tools for pre-launch analytics as part of the SkyDeck Pad-13 Program.

Beyond work, I enjoy sailing, hiking, playing board games, and all sorts of outdoor adventures. I was the national champion in the Chinese Yachting Association Youth Sailing League in 2017.

If you share any of these interests or simply want to chat, don’t hesitate to connect—my calendar is always up to date! :D