video1.0<iframe src="https://www.loom.com/embed/902299ec49334af88c95b6849ac0c505" frameborder="0" width="1890" height="1417" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14171890Loomhttps://www.loom.com14171890https://cdn.loom.com/sessions/thumbnails/902299ec49334af88c95b6849ac0c505-bf02168492336d2d.gif84.328Hello everyone, I'm Raghav from Formatic Labs. In this video, I share our mission to train reasoning models, like GPT-5, to function as autonomous scientists by utilizing various tool calls for data analysis and research. We've developed agentic heuristics through a hackathon to filter out poor outputs, showcasing only the best results on our new archive site. I encourage researchers to explore these findings and provide feedback to help us refine our work. Your insights will be invaluable as we continue to advance this project.