[smj-edison]

I've tried to use Claude Code with Sonnet 4.5 for implementing a new interpreter, and man is it bad with reference counting. Granted, I'm doing it in Zig, so there's not as much training, but Claude will suggest the most stupid changes. All it does is make the rare case of incorrect reference counting more rare, not fixing the underlying problem. It kept heaping on more and more hacks, until I decided enough is enough and rolled up my sleeves. I still can't tell if it makes me faster, or if I'm faster.

Even when refactoring, it would change all my comments, which is really annoying, as I put a lot of thought into my comments. Plus, the time it took to do each refactoring step was about how long it would take me, and when I do it I get the additional benefit of feeling when I'm repeating code too often.

So, I'm not using it for now, except for isolating bugs. It's addicting having it work on it for me, but I end up feeling disconnected and then something inevitably goes wrong.

[onlyrealcuzzo]

I'm also building a language in Zig!

Good luck!

[smj-edison]

Oh cool! I'd love to hear more. I'm implementing an existing language, Tcl, but I'm working on making it safe to share values between threads, since a project I contribute to[1] uses Tcl for all the scripting, but they have about a 30% overhead with serialization/deserialization between threads, and it doesn't allow for sharing large values without significant overheads. I'm also doing some experiments with heap representation to reduce data indirection, so it's been fun getting to learn how to implement malloc and other low-level primitives I usually take for granted.

[1] folk.computer