{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/d775470e12604cffa5500d0236bbc5c8\" frameborder=\"0\" width=\"1280\" height=\"960\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":960,"width":1280,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":960,"thumbnail_width":1280,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/d775470e12604cffa5500d0236bbc5c8-e873e6f39eb7226f.gif","duration":146.589,"title":"Parivahan Seva AI Agent Orchestrator Overview","description":"This Loom explains a redesigned approach for the Parivahan Seva portal to make fine and challan information easier to access. The speakers say the current site is chaotic with thousands of options, so they built a simple orchestrator that you can ask what caused a fine, and it launches agents to search and summarize results. One agent collects evidence and the output provides a summary of the challan, such as a case dated 26 August involving a dispute due to a misread GH to CH. They mention a related email contact for evidence, open AI API usage, alerts around Jalan, and that the demo uses mock-up data."}