I'm also a fish fan and largely agree--but if you are forced to use Windows and you don't care for WSL, nushell is likely your best option. It's pretty good! Almost feels like Unix but you're in Windows. I don't think fish shell will be ported over to Windows anytime soon...
Git works pretty well for syncing binary files. You can even do this:
$ cat .gitattributes
*.db diff=sqlite3
And you can know how the sqlite files changed between commits--but yeah if you make commits while a process is writing new rows or updating data then you'll have data loss if you make a commit on another machine--better to use something like litestream in that case
Some hallucinations stay in codebases longer than others! If there were zero hallucinations there would really be no novel output. Some hallucinations are useful and some are not.
The term "hallucinations" has always frustrated me. The marketing there makes sense, but an LLM that hallucinates is an LLM doing exactly what it was designed for -predicting what a human might say in response.
Facts don't really play a part there, if a response is factual its only a sign that the training set largely agreed on the facts (meaning the correlation of token sequence was high).
