Have you read the "Whispering earring" essay? I love it for the LLM era.[1]

You can treat AI as a whispering earring - "What should we do now? How do we fix this? What do you think?" Or you can treat it like an exoskelton - "Implement kd-tree with metric space xyz for this problem, mapping this to that blah blah".

That's pre-thought execution automation that makes review much simpler - you already know the shape of the desired output. The whispering earring is atrophy.

1. https://croissanthology.com/earring