Free mini-course
Prompt engineering, one task at a time.
A free mini-course for using AI at work: clear requests, usable results and practical checks. Six lessons, about an hour of reading and practice.
Editorial team: Take Two ·
Start lesson 1 ↗What you will learn
You will practise turning a vague request into a clear task, providing relevant information, choosing a usable format and checking whether the answer meets your requirements. Examples cover emails, project notes and support requests.
This is for professionals, managers and operational teams starting to use AI. You need a tool with a text chat and time to try the exercises. Work through the lessons in order, or revisit a topic when you need it.
Prompt engineering helps define requests; it does not remove errors or provide information the system cannot access. The course builds a working method, and includes no certificate.
Six lessons to try at your own pace
- 01
How to write an effective prompt
Start with a concrete task: goal, audience, constraints and success criteria. Includes a complete business email example.
8 min of reading and practice ↗ - 02
Give AI context and constraints
Provide relevant information, separate instructions from documents and handle missing facts. Practise with a project update.
10 min of reading and practice ↗ - 03
Get answers in the right format
Turn notes into a task table with owners, deadlines and missing information. Choose a format that makes the result usable.
8 min of reading and practice ↗ - 04
Use examples to guide AI
Show how to classify requests with input and output examples. Handle ambiguous cases without forcing an answer.
10 min of reading and practice ↗ - 05
Improve a prompt through testing and feedback
Fix one problem at a time and compare versions on normal, incomplete and ambiguous cases. A simple way to evaluate answers.
10 min of reading and practice ↗ - 06
Verify answers and reuse prompts
Check facts and sources, recognise AI limitations and build a reusable prompt template. Includes a final exercise with a commented answer.
14 min of reading and practice ↗
Sources and further reading
The exercises are original Take Two teaching examples. For technical background on instructions, examples and evaluation criteria, see: