There is something very wrong now with how companies operate in general.

You get beaten down eventually. Late last year. I spent like an hour going through why a PR (and this developer's work) in general wasn't acceptable to my superior. He said to me that he was perfectly fine with someone not understanding basic language features (after 6 months using the langauge). He then merged it.

It didn't work (as I had warned) and created a situation where I had to turn off tests in some projects as it totally broke them. I've spent months fixing his crap and still haven't recovered from one bad PR. Now add two other employees that are like this and my manager does nothing about it. I bought a AI package from Jetbrains and now have it do almost all the work. I normally spend some time cleaning it up. Management have made it clear to me that they don't care about quality, they won't hold anyone accountable and won't even fire people that clearly cannot program.

I am 43 years old this year. I just can't be bothered trying to be a hero anymore.

Similarly, my father who retired last week was a joiner/carpenter and would be considered a master boat builder. When my sister was little my dad made her new bed with hearts and flowers carved in the headboard.

He described how adversarial he was too his employer before he retired. He was engaging in Malicious compliance (he is a layman and didn't know it was called that) because management was making his life miserable by employing the same sort of the stand-up meeting ceremony nonsense in carpentry.

They managed to make someone with that level of skill hate their job because of process.

JavaScript in the late 90s was doing a hell of a lot less than it is today.

Yes, languages with very strong type systems like Rust are incredible when paired with an LLM. Just like chat bots have a calculator as a "tool" because they are not the best at calculation themselves, they need a type system to deterministically validate the safety and cohesion of the code they generate.

Its languages like C that you have to watch out for, because the LLM will gladly say "this is safe!" when its not.

Why is that less realistic than saying 'rewrite in rust, make sure there are no memory leaks'?

My point, which I should have been clearer with, is that we aren't at a state where you can just one shot a rewrite of a complex application into another language and expect some sort of free savings. Once we are at that state, and it's good enough to pull it off, why wouldn't the AI be able to pull it off in C as well?

LLMs are great at C, probably because C is historically the most popular language in the world, by far. It only declined slightly very recently. But there's insane amount of code written in it.

Have you tried asking Claude 4.6 Opus?

Based on a FIDO2 spec I used it to write a reasonably compliant security token implementation that runs on top of Linux USB gadget subsystem (xcept for attestation, because that's completely useless anyway). It also extracted tests from an messy proprietary electron based compliance testsuite that FIDO alliance uses and rewrote them in clean and much more understandable C without a shitton of dependencies that electron mess uses. Without any dependencies but openssl libcrypto, for that matter.

In like 4 hours. (and most of that was me copy pasting things around to feed it reasonable chunks of information, feature by feature)

It also wrote a real-time passive DTLS-SRTP decryptor in C in like 1 hour total based on just the DTLS-SRTP RFC and a sample code of how I write suckless things in C.

I mean people can believe whatever they want. But I believe LLMs can write a reasonably fine C.

I believe that coding LLMs are particularly nice for people who are into C and suckless.