Explore OpenAI's cutting-edge approach to reinforcement fine-tuning for code models in this 17-minute conference talk by Will Hang and Cathy Zhou from OpenAI. Gain deep insights into the methodologies and techniques used to enhance AI agents through reinforcement learning, specifically focusing on how these approaches improve code generation and understanding capabilities. Learn from OpenAI researchers as they share their expertise on agent reinforcement fine-tuning, covering the theoretical foundations and practical implementations that drive improvements in AI coding assistants and related applications.