[Zigurd]

It's probably too much inside baseball to merit a study, but I'm curious if the results would change for part-time coders. When I'm not coding, I'm writing patents, doing technical competitive analysis, team building, etc.

My theory is that if you're not full-time coding, it's hard harder to remember the boiler plate and obligatory code entailed by different SDKs for different modules. That's where the documentation reading time goes, and what slows down debugging. That's where agent assisted coding helps me the most.

[nyrikki]

SDKs and Binary format descriptors are where I see agents failing the most, they are typically acceptable for the happy path but fail at the edge cases.

As an example I have been fighting with agents re-writing or removing guard clauses and structs when dealing with Mach-o fat archives this week, I finally had to break the parsing out into an external module and completely remove the ability for them to see anything inside that code.

I get the convenience for prototyping and throwaway code, but the problem is when you don’t have enough experience with the quirks to know something is wrong.

It will be code debt if one doesn’t understand the core domain. That is the problem with the confidence and surface level competence of these models that we need to develop methods for controlling.

Writing code is rarely the problem with programming in general, correctness and domain needs are the hard parts.

I hope we find a balance between gaining value from these tools while not just producing a pile of fragile abandonware