A practical prompt engineering foundation
This mini-course covers the ten core techniques that account for most of the gap between a vague prompt and a reliable one. Each lesson names one technique, explains when to use it, and gives a concrete example you can run. The navigator below lets you step through all ten lessons with their examples. Work through them in order — the later techniques assume the earlier ones.
How the techniques fit together
The first lessons are about phrasing: being specific, showing examples, and asking for reasoning. The middle lessons are about structure: assigning a role, using delimiters to separate instructions from data, and pinning down the output format. The final lessons are about discipline: using negative space to say what you do not want, and refining iteratively. In practice you stack several techniques in one prompt — a role, delimited input, a few examples, and an explicit output schema — rather than using them in isolation.
Tips for applying the course
Run each example against a weaker phrasing so you feel the difference rather than just reading about it. Change one variable at a time when refining, so you know what actually moved the result. Keep a personal library of prompts that worked, with a note on the task and the technique. Once the ten fundamentals are second nature, move on to the advanced techniques — tree of thought, ReAct, and reflection — which build directly on this base.