The FAANG paradox : these companies all have this performance-intensive culture with the gruelling interview, "perf cycles", promotion processes, OKRs, etc... to ensure all engineers function at the top of their capacity. At the same time, their new public facing products have been mostly underwhelming over the last decade. It seems most of the innovation occurred back when FAANGs were 10-100x smaller. (The AI work i…
This I think is a bit like the "IQ selection" phenomenon, where if we suppose IQ actually measures (say,) test-taking ability rather than some complex notion of intelligence... then we find places that try to filter on IQ (law schools, med schools, FAANG) end up selecting for uncreative kinds of intelligence. As a bold, perhaps wrong hypothesis, this might be in play here. We have the very best leet-coders, but the v…
On the other hand if it turns out that you are actually filtering by "willingness to take years of testing" then you might get different results