Experience
2019 -- Present
Senior AI Research Scientist
Meta AI (FAIR)Menlo Park, CA
- Lead a team of 8 researchers working on large language model alignment and multimodal reasoning
- Co-led development of Meta's open-source RLHF framework, adopted by 200+ research groups worldwide
- Published 15+ papers at top venues (NeurIPS, ICML, CVPR, ACL) with 3,000+ citations
- Contributed to the training and evaluation of LLaMA model family
2015 -- 2019
Research Scientist
Google BrainMountain View, CA
- Developed novel reinforcement learning algorithms for robotic control and sim-to-real transfer
- First-authored 8 papers at NeurIPS, ICML, and ICLR on deep RL and representation learning
- Built and open-sourced a benchmark suite for evaluating RL generalization, used by 50+ labs
- Mentored 5 research interns, 3 of whom went on to tenure-track faculty positions