{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/6fd9ed039cf647fdb4422a511b966851\" 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/6fd9ed039cf647fdb4422a511b966851-cc39e3fc8ee07aa1.gif","duration":757.521,"title":"Codepath Final Project ","description":"Hi, I am Ashatu, and I presented my Mood Changer, a self evaluating, self refining, agentic music recommendation system. I ran Python AgentMain.py with three preset vibes and two interactive modes, and it produced top five recommendations, like NotRainyWindow by Paper Lantern. The workflow is plan, act, evaluate, refine, where confidence starts around 0.57 and refinement updates genre and energy until confidence clears a 60s threshold. I also ran pytest, and all 17 tests passed, with fast coverage from 5 percent to 100 percent. No action was requested from viewers."}