video1.0<iframe src="https://www.loom.com/embed/f885f1d2403047cf9d584fd98a1375dd" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/f885f1d2403047cf9d584fd98a1375dd-3fab05023d0ccf19.gif233.22In this Loom I walk through VibeFinder, my final project music recommendation app. It takes a user’s genre, mood, and energy, then scores 20 songs using signals like genre match, smoothness, and energy closeness, and returns the top candidates with a plain English explanation of ranking. I demo Gargoyle conflict detection, where an adversarial profile scores only 3.76 out of 8 versus 7.8 for a low fi profile. I also run an automated reliability evaluator across 5 preset profiles, with 5 out of 5 passing, plus consistent checks showing identical output order. No action is requested from viewers.