Looking at LLMs as a less-than-completely-reliable compiler is a good idea, but it's misleading to think of them as natural-language-to-implementation compiler because they are actually an anything-to-anything compiler.
If you don't like the results or the process, you have to switch targets or add new intermediates. For example instead of doing description -> implementation, do description -> spec -> plan -> implementation
If you don't like the results or the process, you have to switch targets or add new intermediates. For example instead of doing description -> implementation, do description -> spec -> plan -> implementation