Comment by sivaragavan
7 days ago
Aah.. I have been repeating this to my team for several days now. Thanks for putting this together.
I start every piece of work, green or brown, with a markdown file that often contains my plan, task breakdown, data models (including key fields), API / function details, and sample responses.
For the tool part, though, I took a slightly different approach. I decided to use Rust primarily for all my projects, as the compile-time checks are a great way to ensure the correctness of the generated code. I have noticed many more errors are detected in AI-generated Rust code than in any other language. I am happy about it because these are errors that I would have missed in other languages.
> I have noticed many more errors are detected in AI-generated Rust code than in any other language.
Is that because the Rust compiler is just a very strong guardrail? Sounds like it could work well for Swift too. If only xcodebuild were less of a pain for big projects.
Yes, exactly. AI isn’t gonna check things like memory management, data type overflows and such, during generation. It would be great if we catch them at compile time.
Regarding swift, totally hear you :) Also I haven’t tried generating swift code - wondering how well that would be trained as there are fewer open source codebases for that.