{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/4ac3a0d3d4774b1794cd01a3404efc09\" frameborder=\"0\" width=\"1728\" height=\"1296\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1296,"width":1728,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1296,"thumbnail_width":1728,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/4ac3a0d3d4774b1794cd01a3404efc09-e1afb98ccbf0030b.gif","duration":466.562,"title":"VibeFinder AI Music Recommender Project Demo","description":"Hi, I am Akash Talota and this is my final project, VibeFinder, an end to end AI music recommender. I show the architecture and code flow, classify, retrieve with RAG genre context, plan structured preferences, act with an explainable deterministic scoring engine, and reflect with confidence validation and bounded retrieval attempts up to 2. I also run three live demos, including chill, Arctic Monkeys for indie rock, and high energy hip hop for the gym, with ranked top 5 songs and an agent timeline. Finally, I cover reliability evaluation using an evaluation harness with pass or fail constraints. I did not ask viewers to take any action."}