Brandon Cui

Brandon Cui is an AI and machine learning researcher at NVIDIA whose work focuses on reinforcement learning, large language model training, and multimodal reasoning. He holds BS and MS degrees in computer science from Stanford University and has previously held research positions at Facebook AI Research (FAIR) and Databricks Mosaic AI Research. He is known for contributions to multi-agent reinforcement learning, including off-belief learning and k-level reasoning for zero-shot coordination, as well as recent work on introspective training methods for LLMs and dense language annotation for robot policy learning.