OATML

Sagalpreet is a DPhil student in the OATML group at the University of Oxford, supervised by Yarin Gal. His research focuses on advancing continual learning methods for foundation models to improve adaptability, memory retention, and efficiency in evolving environments. He completed his undergraduate degree at IIT Ropar. Prior to Oxford, he was a pre-doctoral researcher at Google DeepMind, contributing to Gemini Thinking, continual learning, and recursive self-improvement (RSI) efforts, with an emphasis on reinforcement learning and agents. He also held engineering roles at proximal.ai and Oracle’s speech AI team. His past research spans RL, multi-agent systems, and domain adaptation, with publications at venues including UAI, AAMAS, and RLC. His studies are supported by an LASR studentship.