video1.0<iframe src="https://www.loom.com/embed/99985f6fb7af4546873ad6b57916f724" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/99985f6fb7af4546873ad6b57916f724-fcc7146f431a7d77.gif181.746This is my final project for Apply AI, a music recommendation system that matches songs by genre, mood, and energy. I show demos where I run python -m src.main, load a dataset of 18 songs, and enter requests like chill, pop, and why these songs. The router acts like an agent to rank the songs and explain why each one was selected based on energy and mood similarity. For a random, unclear request, the guardrails help the system respond safely. I do not ask viewers to take any action.