{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/bf0b02975e4e4b3ba93929199c747481\" frameborder=\"0\" width=\"1920\" height=\"1440\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1440,"width":1920,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1440,"thumbnail_width":1920,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/bf0b02975e4e4b3ba93929199c747481-c9bb3538cecdcb8b.gif","duration":424.661,"title":"Enhancing Music Recommendations with AI RAG","description":"Hi, I am Guaymara, and this is my AI110 Section 1B final project on enhancing a music recommendation system so every recommendation decision is visible to the end user. I upgraded my original weighted scoring approach using six features to make a score out of 10, including genre, mood, energy, and confidence labels. I added an AI RAG layer with strict instructions to not hallucinate songs, using catalog context to rank results with rationales. I also ran evaluation tests across genres and prompts, and all bar tests are passing. I did not ask for any specific viewer action, but I recommend you watch the demonstration and reach out with questions."}