Skip to content
Lumina

Course bundle · Core + one specialty

Complete · AI evaluations

Build repeatable tests for AI answers, then decide which failures should block a release.

modules
16
Python labs
15

Launch priceUS$240Regular price US$300

Buy on Hotmart Read free lessons

15 days to request a refund through Hotmart after purchase.

Fictional example

Try this decision

  1. What happenedAn assistant gets a high score after guaranteeing a refund that the policy only offers to review.
  2. What to testSave the request, the answer and the source. Write a test that checks whether the source supports the claim.
  3. What to recordPut the case in your test set and decide whether this kind of failure should block a release.

Who it’s for

For engineers and QA practitioners who need to compare an AI application against the same cases before and after a change.

What you get

  1. Core · The method

    9 modules · 8 optional Python labs

    Recognize the failure, choose a check and explain the release decision. Work through the main exercises without writing code.

  2. Ship LLM Apps on Evidence

    7 modules · 7 Python labs

    Apply the same method to this specialty, with its own reading and offline Python labs.

How it works

Reading and practice
After payment, download the bilingual ZIP from your Hotmart library. Extract it on a computer and open START-HERE.html in your browser.
What you need
A computer and browser for reading. Use Python 3 and a terminal for the labs.
Price and currency
The price is in US dollars. Hotmart shows the final currency and total before payment.
License and updates
Personal license for the downloaded edition, with 12 months of package updates.
Buy on Hotmart