Guardrails in Practice
Add guardrails to an AI help assistant in real code: a check on what comes in, a check on what goes out, rules from a one-page AI use policy, and a step where a person approves every refund. Then measure what each one catches and wrongly blocks.
Modules
6
Verified
0
In Progress
0
What you walk away with
- ✓A shareable portfolio URL with your project walkthrough
- ✓Module-by-module timeline of everything you built
- ✓All the code — pipelines, guardrails, deployment configs
- ✓Production patterns documented on your profile
An intermediate build course for developers and AI product owners. You work on Ask Corri, the help assistant of a made-up home insurer, in a small Node app that runs with no API key. You score an input check on 200 labelled customer messages and see that a stricter check catches more problems and blocks more real customers, build an output check for personal data, made-up cover names and promises on 30 draft answers, turn a one-page policy into rules with test cases, add an approval step so no refund is paid without a person, and finish with a report on what each guardrail catches, wrongly blocks and costs in time. Educational, not legal advice.