[mixmastamyk]

I don’t use them, yet. Am open to it, but no longer trust most tech companies. Probably an open model in several years on a beefy yet affordable machine I’ve yet to purchase.

Also am at the peak of my game, and automated templates, snippets, and stackoverflow lookup a decade+ ago. I prefer reading a discussion of tradeoffs to approaches before picking one. It may take up to ten more minutes up front but save hours later.

So waiting for the dust to settle.

[TimorousBestie]

Work kind of guilt-tripped me into giving it a shot and my experience with Claude was such a time-waster that I’ve kinda been ignoring it and continuing to do my own thing.

Maybe other people are better at prompting or designing VSCode integrations but what I’ve experienced so far has been a mess. Utterly nonsensical design decisions, it doesn’t seem to understand basic linear algebra or the LAPACK API. (I tried adding the Fortran or C source to its context to no avail.) I asked it to rewrite a well-documented scalar function using AVX intrinsics and. . . woof. No good.

Hopefully the field either improves dramatically in a couple years or goes back into hibernation for the next AI winter.

[TheCleric]

Same. I gave up after 30 minutes of correcting it when it kept suggesting code to use that was completely invalid.

[herbst]

There are several companies with proper privacy terms offering available models as pay per use to a fair price.