Learn the whole method, one calm step at a time.

We build your first harness together, on screen, in plain English. Nothing is assumed, nothing is rushed. By the end you will have a working system and the confidence to use it on a real evaluation.

What you'll be able to do

By the end of the course

Recognise an unsafe claim

You'll spot the sentence that looks fine but has no source behind it, in seconds.

Set up your own rules

Copy our instructions into your AI, and adjust them for your own donors and templates.

Check any report

Use a one-page checklist to review a whole document before you sign it.

Keep client data safe

Know exactly what to share, what to remove, and what to write in a disclosure note.

The syllabus

Six modules, explained from the beginning.

#	Module	What you'll understand
1	Why AI gets things wrong	What a “hallucination” is, why it happens, and why confident wording is not evidence.
2	The seven costly mistakes	Each mistake shown in a real paragraph, with the fix right beside it.
3	Building your harness	The four files, written with you on screen: the Skill, the test set, the checklist, the proof table.
4	Applying it to real work	Logframes, indicators, reports, interviews and data, step by step.
5	Privacy and donor rules	Keeping client data safe, and the exact wording to disclose your use of AI.
6	Making it a habit	How to review work consistently, and how to explain your process to a donor.