muddassirsayyed/ai-music-recommendation-via-embeddings-ai-ml-software ? reverse-engineered prompt

Build me a simple hackathon style AI music recommendation app that suggests songs based on what a user likes. I want it to feel like a small working demo where someone can type in a few favorite songs, artists, moods, or genres, and get back a list of similar music recommendations.

Keep it straightforward and practical, with a clean interface and a basic backend that makes the recommendations using embeddings or some kind of similarity matching. If you need to, look up current docs online for the best way to do the embeddings part. Make the experience easy to use, with a place to enter preferences and a results area that explains why the songs were suggested in plain language.

It does not need to be fancy, but it should actually work end to end and look polished enough for a hackathon demo. If any data is needed, use a simple sample dataset or a lightweight approach that can be run locally.