Interesting. Aside from conversion of tracks to midi, it looks like i could run my music collection through this and use the output to create a recommendation system (along with existing metadata and lyrics)
I thought it would be the opposite, because you can't say: "The AI was trained on a bazillion lines of idiomatic Nim 3 solving a wide range of common problems"
Yep that is the expectation one could have but it is not the experience I (and many others) had.
The explanation I have is that it is so close to pseudocode that it only needs to know just a tiny bit about Nim peculiarities (not too many) to do a good job (plus static types and compiler help a lot).
Ai writes nim pretty well. Its close enough to other languages that llms can manage to write performant code with few issues. There are some gotchas, but the fast compile times and good error messages makes it easy to iterate fast for the bots.
I dunno if you're right, or if it matters. The reply is decent. It's exactly what I'd do (or "am doing"). Running a model locally, tweaking configs, experimenting with prompt etc. Not only have I learned a ton about running models but I now have a useful personal AI assistant, researcher and code reviewer who I chat to over Telegram and send issues to on github.
Spf, DMARC, DKIM all in place and my wife still has her own-domain outbound mail blocked by a government org who use some shitty spam protection that ignores all of that, blocks our email provider and doesn't even allow her to reply to emails they sent her.
True story: two college friends of mine having a conversation on Friday evening, Friend A (single) asked Friend B (in a relationship) what he was doing on Saturday. B's excited reply was "Deus Ex!"
reply