{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/cb404a9d925046508916a9cf0bd8bd7a\" frameborder=\"0\" width=\"1664\" height=\"1248\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1248,"width":1664,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1248,"thumbnail_width":1664,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/cb404a9d925046508916a9cf0bd8bd7a-28cb1a3c1e89e78a.gif","duration":271.3819,"title":"Demonstrating Parallel Research Agents for Kiroween [video2]","description":"In this video, I demonstrate the functionality of my final Python script, which utilizes a parallel research agent architecture to assess compliance for an auto loan approval system under EU regulations. We ran five sequential and parallel subagents, achieving a risk classification of high with a score of 65 out of 100 and a confidence level of 95%. The relevant articles identified were six, eight, and nine, along with documented compliance gaps. I also showcased a visual representation of the results and ran additional scenarios, including prohibited and minimal risk systems. Please take a look at the results and let me know if you have any questions or need further clarification."}