About me
Hi, I am an M.S. student studying computer science at Carnegie Mellon University (CMU). Before that, I graduated from The University of Hong Kong (HKU). I am passionate about RL Infrastructure, and I am a contributor to verl, slime, Miles, and SGLang. I am currently a Research Engineer Intern at ByteDance Seed, where I work on verl and verl-uniagent. Previously, I was fortunate to work with Prof. Tao Yu at XLANG lab on computer-use agents (CUA) and multimodal GUI grounding, and interned as a machine learning engineer at Apple China.
Open to New Graduate Opportunities I am looking for new graduate roles and will graduate in December 2026. Please reach out if your team is hiring.
Experience
Open Source Projects
My work includes, but is not limited to View 7 pull requests
- verl #6974 Sharded delta weight sync over NCCL for disaggregated rollout
- verl #6804 Add multimodal continuous tokens
- verl #6779 Add continuous tokens for agentic rollout
- SGLang #19225 Support Stable Diffusion 3 Medium with the SGLang backend
- SGLang #20464 Add the update-weight-from-tensor pipeline to diffusion
- SGLang #22817 Extract post-training weight APIs into mixins
- slime #1141 End-to-end Geo3K multi-turn RL recipe for VLMs
Professional Experience
-
ByteDance Seed
Research Engineer Intern Β· May 2026 β Present Β· San Jose
Working on RL infrastructure for agentic RL β contributing to verl and verl-uniagent.
-
XLANG Lab
Research Assistant
Working on computer-use agents, GUI grounding, etc.
-
Apple Inc.
Machine Learning Engineer Intern
Working on RAG-based Q&A system.
-
UC Berkeley CDSS
Research Assistant
Working on natural language processing and digital humanities.
Publications
-
OpenCUA: Open Foundations for Computer-Use Agents
NeurIPS 2025 Spotlight (3%) & COLM 2025 Workshop Best Paper
