I see it more as “this result is commensurate with my experience,” the problem with it, of course, being reliant on the pilot’s experience. But otherwise we’re

I see it more as “this result is commensurate with my experience,” the problem with it, of course, being reliant on the pilot’s experience. But otherwise we’re just talking about a “sniff test” on a number. LGTM sucks in software because software inherently is a lot more complex than a number.