>there is no sign of the AI slowing down or being unable to maintain it
This means you designed the system well and made sure the agent implemented it correctly. Even at Big Company the big shot architects don't review every line of code, and nobody seemed to mind that
I'd like to think it comes down to my expertise, but I don't believe I did a lot of designing.
Sure, I've given a lot of inputs over the time, probably I've steered it to solutions that work.
Maybe it was also helpful that I set up a skill to use a different model to review the code for correctness, technical debt, and whether the location where the code is added is appropriate or warrants a new file or abstraction. I set this up long before subagents were widely used.
It’s just a fact that an ai will write and read faster than me. I can’t compete so getting ai to do that has freed up both time and mental capacity to actually think about the implementation details.