{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/1d0925a7d75342cb8b1d78569ecf9723\" frameborder=\"0\" width=\"1500\" height=\"1125\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1125,"width":1500,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1125,"thumbnail_width":1500,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/1d0925a7d75342cb8b1d78569ecf9723-d6af62b21bb11af0.gif","duration":503.18,"title":"SoundFinder 9001 Music Recommender Enhancements","description":"This is SoundFinder 9001, my Applied AI Systems final project, extending the Module 3 Music Recommender. I added an AI-assisted Profile Builder, a RAG grounded explanation using a knowledge base, and a Reliability Harness that runs the full pipeline. I built a study profile for rock acoustic guitar, focus mood, and generated five recommendations with score breakdowns and a top pick like Drift Beacons. The harness ran in about 5 minutes, used five fresh personas, made about 20 API calls per run, and cached results. No action is requested from you, but you can try different profiles and song counts."}