“The best prompt engineers are meticulous: they constantly run experiments, they make detailed notes on what works and what doesn’t, they iterate on their prompts and try to figure out exactly which components are necessary for the prompt to work and which are just a waste of tokens.
Being able to resist superstitious thinking is really important. Like LLMs, humans are pattern matching machines! It’s easy to come up with a prompt, get the intended result and learn entirely the wrong lessons from it.”
