Post #2637536
2026-02-14 22:22 UTC
If I had to rate this article on a propaganda scale, I'd give it a 3. Anthropic is absolutely engaging with how hard it is to get serious code out of LLMs. Just the fact that there is a known-good implementation (gcc) that the agent was allowed to reference means it isn't relevant to 75% of the work I do. BUT THAT'S OK.
Point your over-eager boss to this article, explaining how limited this problem really is, and how weakly it was solved, and at what cost. This is the state of the art today.
Replies (1)
-
@cocoaphony@mastodon.social 2026-02-14 22:31
But I also *do* believe this kind of setup will eventually (maybe 5 years, but I personally think more like 10-20) become a normal part of software development, in the same way that compilers have replaced hand-coded machine language. A next-generation compiler, that still requires quite a lot of engineering to make it work, but erases certain kinds of problems. And so I'm very happy to see Anthropic explore this space to figure out how that might work, even if I think it's miles from shipping.