I spent 18 months rebuilding my algorithmic trading platform in Rust

I've done a mid-size CLI tool with a local server (Axum) at work last year. While I'm happy how it turned out and don't have regrets like OP, I can understand his frustration. Rust is still overhyped, not as much as some years ago, but from all use-cases out there many would have been better done with a GC language. The amount of energy and time spent on solving the language problems instead of the project's goals, isn't payed back for "standard" applications. It's not that your C# or Go services explode every week or are too slow.

There're many excellent use-case for Rust. Just not as many as parts of the "internet" suggests.

Regarding error handling: Even with RUST_BACKTRACE=full, it's most of the time too hard to find the source of the error.

I do not understand taking so long to arrive at these conclusions. Each of them smacks you in the face pretty quickly if they bother you, so it's strange to invest so much time before deciding they are rant-worthy.

Writing web servers is not Rust's primary use case, and you had better have a darn good reason to choose it over dedicated server languages. I laughed out loud when he said, "Just give me a garbage collector..."

Use. Python. You ain't writing an operating system or low-latency data streaming engine. You don't need Rust.