@nprofile1q... fair point, perhaps, it is my own pattern of learning I projected here. I know that personally, I wont be able to spot an LLM mistake on a specific algorithm even if I read about it in a book 4 years ago. If I coded such an algorithm, then the likelyhood of me spotting a mistake would be much higher.