Every coding assistant or LLM I've used generally makes a real hash of TypeScript's types, so I'm a little skeptical, but also: > RustAssistant is able to achi

Every coding assistant or LLM I've used generally makes a real hash of TypeScript's types, so I'm a little skeptical, but also: > RustAssistant is able to achieve an impressive peak accuracy of roughly 74% on real-world compilation errors in popular open-source Rust repositories. 74% feels like it would be just the right amount that people would keep hitting "retry" without thinking about the error at all. I've found LLMs great for throwing together simple scripts in languages I just don't know or can't lookup the syntax for, but I'm still struggling to get serious work out of them in languages I know well where I'm trying to do anything vaguely complicated. Worse, they often produce plausible code that does something in a weird or suboptimal way. Tests that don't actually really test anything, or more subtle but actual bugs in logic, that you wouldn't write yourself, but need to be very on the ball to catch in code you're reviewing.