About Me

I’m an AI research scientist at Meta Superintelligence Labs, in the Fundamental AI Research team (FAIR). I work on Large Language Models (LLMs) and AI Agents. My current focuses include:

• Frontier Agentic LLM Development: I contribute to mid-/post-training of Meta’s Muse model family, focusing on agentic reinforcement learning for coding (e.g. SWE-Bench, Terminal Bench), automated research (e.g. MLE-Bench), and OpenClaw-style tool-use tasks (e.g. Claw-Eval, ClawBench).

• Fundamental LLM Research: I also lead and conduct research projects covering topics like long-horizon agentic reinforcement learning, multimodal reasoning, safety post-training, and hallucination mitigation.

I received my PhD in Computer Science from the University of Toronto, where I studied automated understanding and generation of creative language use. During my PhD, I was fortunate to intern at Meta FAIR and Google Research.