[layer8]

What’s interesting to me is reasoning about the problem and its implementation. And that doesn’t stop at any abstraction level. Reasoning in the small is just as important as reasoning in the large. And the issue with LLMs is that their capacity for sound reasoning is limited. They are sloppy on any level. You can’t get them to be thorough and dependable in reasoning, regardless of the abstraction level.

[gitaarik]

Well I think the reasoning of coding agents on lower levels is good enough for me that I don't have to constantly be involved with it, only occasionally have to dive in and help out.

[layer8]

I don’t think that the logical reasoning ability of LLMs depends on the abstraction level. Their heuristic knowledge differs between levels, but that’s a different thing. My concern is the reasoning capabilities.

[gitaarik]

And so you experience that AI generated code, even on lower levels, is not good enough for you to be more productive?

[layer8]

It may be good enough to make me more productive, but only because I don’t relent on ensuring that the code is well-reasoned. Indeed, I don’t experience that when I do relent.

[gitaarik]

Ok, so you're basically saying that the AI-generated code does it's job, but when you actually review it you think the way it does it's job is not as it should, and if you get into that flow, your agentic productivity goes down?