Skip to content
How to talk to AI

Prompt engineering for beginners

The short version
A vague prompt ('help with science') produces a vague reply. A specific prompt names the goal, the audience, the constraints, and what good looks like. Prompt engineering is not a secret handshake. It is writing the assignment clearly.
The five parts
Goal: what should exist when you are done. Audience: who will read it, and at what age. Constraints: length, format, things to avoid. Evidence: examples of the pattern you want. Check: how you will tell a good answer from a fluent one. Skip any of those and the model will fill the gap with averages.
Try this
Write v1: 'Explain photosynthesis.' Write v2: 'Explain photosynthesis to a curious 9-year-old in eight sentences. Do not mention graduate-level chemistry. End with one question the child can test with a plant on a windowsill.' Run both. Score them on accuracy, age fit, and whether the last line is actually testable.
What WonderMind trains
Prompt Lab is built around that loop. The lesson is not 'the magic words.' The lesson is: you stay the editor. If the output is wrong, you change the brief, not your standards.

Turn reading into a test

Create a free learner profile to run the experiment and keep your evidence in a private archive.

Start learning →