I -have- built something like this for the TV using NFC cards, which was a great first-electronics-project for myself. That said, the most frustrating part is not the actual hardware itself but getting whatever streamer you're using to play the content you want. For example, this project required the author to WireShark and reverse engineer how Chromecast managed things.
If you do go down this route, I found that Plex offered the best deep-linking functionality and would wrap all of your content with that... but it was still somewhat unreliable.
To this point, I've been using Qobuz as an alternative and it's recommendation engine is laughably bad, but the experience is somehow better. I'll get the most random songs pop up in the list, and sometimes it's a very pleasant surprise.
In the world of music discovery a bad recommendation engine is maybe better than a hyper-fine-tuned one.
I've been working on a small side project on-and-off for the last year and some. The state-of-the-art decisions I made last year are already dead. There's a new version of everything, and everything was a breaking change, nothing is backwards compatible.
Claude 3.5 is a trusted developer partner that will work with you and outline what it’s thinking. It’s not always right but because it outlines its reasoning you too can reason about the problem and catch it.
ChatGPT, for me, was a stack overflow solution dump. It gives me an answer that probably could work but it’s difficult for me to reason about why I want to do it that way.
Truthfully this probably boils down to prompting but Claude’s out of the box experience is fantastic for development. Ultimately I just want to code, not be a prompt wizard.
If you do go down this route, I found that Plex offered the best deep-linking functionality and would wrap all of your content with that... but it was still somewhat unreliable.