{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/5453f9dd4b89407d93a0e08428267941\" frameborder=\"0\" width=\"1660\" height=\"1245\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1245,"width":1660,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1245,"thumbnail_width":1660,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/5453f9dd4b89407d93a0e08428267941-8a94897e707b8369.gif","duration":231.169,"title":"Revolutionizing Clinical AI: Training Agents to Assist Physicians","description":"In this video, I discuss the pressing issue of physician burnout, primarily driven by lost time to HR systems. We are developing a residency-like environment for clinical AI agents, allowing them to learn in an IRBA setting. Unlike traditional question-answering tasks, these agents must integrate multiple data points and reason across them, which requires reinforcement learning rather than a simple API call. We also address privacy concerns and the need for medical reasoning by simulating real-world interactions with fire servers. My goal is to create a small, efficient model that assists doctors, freeing them to focus on what only they can do."}