[ModernMech]

I've been doing this recently; I don't really know swift, and I wanted to see how well I could use almost 100% LLM to do my coding. For the most part, I have been able to get a pretty substantial app going, but once it gets to a certain size, things start to become hairy.

For instance, I am adding bluetooth capabilities to my app, and the bluetooth manager can be quite large, to the point where the LLM context window starts to become an issue.

When it gets to that point, the LLM can start making suggestions that don't make sense; it's starts forgetting what you had already done, it can start making architectural suggestions that contradict what you had been doing before. So if you follow them, you end up in this weird state mired in conflicting decisions. Almost like design by committee without the committee.

It's important to recognize this when it's happening rather than just blindly following what it suggests, so some expertise at engineering is still necessary even though the LLM can write okay Swift code.

[amelius]

Maybe you should ask the LLM to write concise documentation for the functions that it generates (or ask it afterwards). Then use that documentation in the context-window instead of the actual, more lengthy code. Compartmentalize your code into modules/functions with a clearly defined boundary (api).