I think the workflow of translating English to Python, working in Python, and then translating Python to Rust sounds a lot more error prone than just learning R

I think the workflow of translating English to Python, working in Python, and then translating Python to Rust sounds a lot more error prone than just learning Rust if that's your target language. Especially if other people are using different languages as their intermediate language. Idioms are different, best practices are different, patterns are different, and no LLM is going to be able to seamlessly go back and forth while maintaining perfect fidelity and also maximizing the performance benefit of Rust. And how would engineers even communicate with each other if each is seeing something completely different? Maybe there'd be some opportunity at the small scale, like if you're only a Python person and need to make a small change to a Rust codebase, it might be able to give you some Python pseudocode, and you could maybe make some changes to that and it'd apply them in translation, sure, but you're still going to be on the hook for reviewing the Rust translation; it's not something that can be fully trusted and automated, and it's hard to see where this would scale beyond small code snippets that were easily translatable.