{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/833304bb1feb423aa5a15dd0c38d2be0\" frameborder=\"0\" width=\"1670\" height=\"1252\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1252,"width":1670,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1252,"thumbnail_width":1670,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/833304bb1feb423aa5a15dd0c38d2be0-617a2a0d31be89da.gif","duration":185.184,"title":"YC Hackathon - Dogfight","description":"In this video, I discuss how brainstorming with LLMs can lead to more comprehensive solutions, but often requires multiple iterations to cover all angles. I introduce DOPFITE, our MCP server that allows different LLMs to impersonate experts and debate solutions collaboratively. For example, we explore whether to host performance metrics centrally or per customer. I encourage you to think about the alternatives and propose solutions as we refine our approach."}