Gokul Swamy

Gokul is a final-year PhD student in the Robotics Institute at Carnegie Mellon University, working on efficient algorithms for interactive learning (e.g., imitation/reinforcement learning/RLHF). His research focuses on developing the novel algorithmic paradigms required to build robustly aligned agents that gracefully handle situations unseen in their training data. Gokul designs scalable algorithms for domains like robotic manipulation and language modeling. He's spent summers working on ML at SpaceX, Perception at NVIDIA, Motion Planning at Aurora, World Models at Microsoft, and LLMs at Google.