I'm a Research Lead at Yutori.
I currently work on building Computer-Use Agent (CUA) models through mid-training and reinforcement learning, and most recently have been driving the development of Yutori Navigator and Yutori n1.
Previously, I led post-training for LLaMA 3 Multimodal on the LLaMA team at Meta GenAI and co-led Emu, Meta's text-to-image foundation model.
Research interests: large language models, multimodal learning, AI agents, and reinforcement learning.