Measured

Courses

Measured Academy

by Hassan Mansour

Courses for people who want to build with AI and know whether it works. Every module ends in a test you pass by running real code. The Academy runs on Maktab, a learning platform from the same company, and enrolment will happen there, not on this site.

Launching soonNo course is open yet. There is no enrolment and no price yet.

Measured Academy is for adults. You must be 18 or older to enrol.

How every course works

  1. The first module is free

    When a course opens, its first module is free once you have created a free account (sign-up required, no payment), so you can judge the course first.

  2. A test in every module

    You run real code on your own machine and hand in the file it wrote. A check script tells you whether you passed.

  3. Your own API key

    You use your own key. Each module states what it is expected to cost before you start, and says so plainly where no run has measured it yet.

  4. A Certificate of Participation

    Finishing a course earns a Certificate of Participation. It records that you took part, nothing more.

Course 1Launching soon

AI Engineering, Measured

For working developers who ship software and are new to building with language models.

The book's chapters, one module each. This is the first course that will open.

Written in English. A French version has not been started.

Module map4 of 15 modules written

  1. 1Module 1: Why AI Development Feels DifferentWrittenFree at launch, with sign-up
  2. 2Module 2: Picking Your Stack Without Getting ParalyzedWritten. Its test run on a current model is still to do.
  3. 3Module 3: Structured Output: Getting Data Back, Not TextWritten. Its test run on a current model is still to do.
  4. 4Module 4: Your First Agent: A Loop, a Tool, and a Stop ConditionWritten. Its test run on a current model is still to do.
  5. 5Module 5: RAG: When to Retrieve, and How to Tell It FailedOutlined, not written yet
  6. 6Module 6: Vector Storage: An Index That Outlives the ProcessOutlined, not written yet
  7. 7Module 7: Conversation State: What to Keep and What to DropOutlined, not written yet
  8. 8Module 8: Multi-Agent Systems: When to Split, and When Not ToOutlined, not written yet
  9. 9Module 9: MCP: Using Tools You Did Not WriteOutlined, not written yet
  10. 10Module 10: Evaluation: Catching Regressions in a System That Won't RepeatOutlined, not written yet
  11. 11Module 11: Security: Bounding What the Model Can DoOutlined, not written yet
  12. 12Module 12: Observability: Tracing What Actually HappenedOutlined, not written yet
  13. 13Module 13: Cost: What It Spends, and How to Spend LessOutlined, not written yet
  14. 14Module 14: Streaming, Async, and Jobs: Handling the WaitOutlined, not written yet
  15. 15Module 15: Your Own System: The First Three MovesOutlined, not written yet

Course 2Planned

AI Vision, Measured (working title)

Lessons with a live camera on how a vision model sees, scores, fails and gets measured.

Eight lessons are drafted. The pass with a real camera is still to do, so nothing here is final.

Lessons

  1. 1Lesson 1: What the model sees
  2. 2Lesson 2: Where the score comes from
  3. 3Lesson 3: Why it breaks
  4. 4Lesson 4: Confidence isn't accuracy
  5. 5Lesson 5: Acting on it
  6. 6Lesson 6: Making it better
  7. 7Lesson 7: Same engine, new skin
  8. 8Lesson 8: Build it yourself

Planned after these

Planned
  • Measured Bootcamp: Build an Agentic AI SystemFor developers moving into agentic AI work.
  • Vibe Code a Real AppFor students.
  • AI in Your Daily WorkFor professionals and teams.

Packs for schools and programs are planned. For Institutions

Waitlist

Waitlist opens soon

There is no form here yet, so this page collects nothing about you.