The skill of writing, testing and systematically refining instructions for language models so they give accurate, consistent results.
Prompt engineering is the craft of writing good prompts. It covers setting a role, explaining the task step by step, giving examples of the desired output, specifying a format such as a table or list, and stating clear constraints. There are no magic words, just clarity, experimentation and steady refinement.
In AI-powered apps, a small change in wording can noticeably improve or break the quality of answers. That is why developers treat prompts like code: they write them carefully, test them against real cases and keep track of previous versions so they can roll back.
Practical tip: when you dislike a result, do not rewrite everything. Change one thing at a time, such as adding an example or clarifying a constraint, and compare outputs on the same set of questions so you know what actually made the difference.