I personally haven't figured out why there isn't a tool that just loops the AI severals times on a task by compiling, feeding in the errors adjusting and repeating and then letting the user review the result be in a success or failure by exceed the loop limits.

Claude has a tendency to overcomplicate bug fixes. If you put it on loop it would probably build a custom ERP.

They exist in various forms...

The LLMs want to please, they can infinitely keep making 'important' changes; changing variable names without changing anything else, adding/removing comments, adding removing print statements, moving a function somewhere in the code (usually by first duplicating it) and fixing perfectly fine code with 'let me see if everything is really working' (it was and now it's broken again).

"Wanting to please" is just fine-tuned implicit prompt stuff, though. You can tell the LLM that it's roleplaying as a senior engineer, and should consider you, the human to be a junior engineer it works with, who is asking it for advice. You can just leave it at that (often works, if the AI knows enough about how software is produced); or you can describe particular traits of senior engineers — e.g. "wants code that works" and "knows when to say the code is good enough and that the junior is now wasting time by fiddling with it."

https://syntix.pro automatically executes and verifies the code it generates. Disclaimer: I made it.

Devin does that

Wasn't Devin a scam? Honest question; Google is not very clear about it but some people said it was a scam/money grab. You are saying it works?