In the project I am working on I use 5 LLM models

• 3 evaluate a task
• 1 implements task
• 1 checks the work and send back to the implementer

What I have found is that anything done with a single family of models is going to be, in my opinion, far less quality than something done with at least a second family of models checking the work. In this particular case it also sounds like there was very little human oversight. Even it works, it likely will have lots of decisions which will come back to haunt the developer, or the users of the software.

AI slop that’s been developed in a mere 4 days (?)

I have been working on a project, with AI, for 4 months. When I see something done in 4 days that is supposed to be a large package, even more complex than what I am working on, I tend to be very wary about the quality.