Cyber & AI Security · AI Application Security
Go deeper. Real-world tasks that need more research, planning and a polished result.
Task 1 of 3
A test plan for a chatbot or AI writer.
What to do
Write a quality test plan for an AI feature (a support chatbot or product-description writer): what to test, sample inputs, expected behaviour, and how to score results. Run it.
You’ll make
Done looks like
A test plan with results from running it.
Task 2 of 3
Make a simple AI app, break it, then improve it.
What to do
Build a small AI app on your machine (an FAQ bot for a brand). Try to make it give wrong or off-topic answers, then improve its instructions and checks.
You’ll make
Done looks like
A working app, the problems you found, and the fixes with proof.
Task 3 of 3
What could go wrong and how to prevent it.
What to do
For a support bot that can look up orders and offer refunds, draw how data flows, mark where mistakes could happen, and list the top 10 risks with fixes.
You’ll make
Done looks like
A data-flow diagram and the top 10 risks with fixes.
Learn before you build
Free course
Free courses on AI fluency and building with Claude, with certificates.
Free tool
Open-source tool to test prompts and AI apps against test cases.
Article guide
How to design AI products people understand and trust.
Article guide
Real cases of AI going wrong, useful for review and testing.
Article guide
How to evaluate AI outputs systematically.