Summary
Huanzhi Mao is a PhD student and student researcher at UC Berkeley specializing in AI and ML systems, with a focus on LLM agents and their tool-use capabilities. He leads the Berkeley Function Calling Leaderboard (BFCL) and is lead author on multiple BFCL releases and a co-first author on Gorilla OpenFunction v2, contributing to widely referenced open benchmarks for function-calling in LLMs. With four years of industry and research experience—including an AI research internship at Salesforce and engineering roles at startups—he blends experimental rigor with production-minded engineering. Huanzhi has also led course infrastructure and content as Head TA for CS 61A and founded CalCourse, reflecting a commitment to education and tooling beyond research. Notably, his work ties benchmark design to practical agent behaviors (multi-turn, web search, and memory), surfacing real-world failure modes for next-gen LLM tooling.
4 years of coding experience
4 years of employment as a software developer
Doctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at University of California, Berkeley
High School Diploma, High School Diploma at Brewster Academy
Middle School Diploma, Middle School Diploma at Nanjing Foreign Language School
Chinese, English, French