AI course for your role · 29 chapters · ~25h 59m

Developer

Read and debug code generated by AI, build agent features, and check changes before release.

Start this course View course outline

who this is for

For the software developer on your team. Assign this course to people responsible for this work, whether they are new to AI or already use it. They build a shared foundation, practice in their role studio, and learn the tools needed to read and debug AI-generated code, build agent features, and check changes before release.

what you'll be able to do

Practice how to read and debug AI-generated code, build agent features, and check changes before release. Build work you can explain to a colleague, with checks and a clear point for human review.

your course, in learning order

start with the map, then the early reading labs so names and values are not a surprise. then agent studio: ship a metered, testable agent feature with a real failure boundary. later chapters add the model, tooling, and evaluation skills needed to maintain it. self-paced; the full route is listed below. Chapters are numbered in the order you take them in this course.

  1. 1. Foundations

    Shared AI foundations and the prerequisites for this role.

    01from worker to builder~1h 45m
  2. 02read the script~21m
  3. 03read the data shapes~23m
  4. 04read the failure~24m
  5. 05mutation~10m
  6. 06read the pipeline~26m
  7. 07intro to terminal~26m
  8. 08intro to claude cli~32m
  9. 09intro to openai codex cli~32m

  10. 2. Role practice

    Practice the work of a software developer in the role studio.

    10agent studio~1h 6m

  11. 3. Applied skills

    Build on the studio with tools, workflows, and checks for your role.

    11git and github cli~36m
  12. 12secrets~16m
  13. 13llm apis~45m
  14. 14structured output~38m
  15. 15mcp~40m
  16. 16agent loops~1h 11m
  17. 17prompting ai coding tools~59m
  18. 18reading agent traces and telemetry~30m
  19. 19test the agent before you trust it~57m
  20. 20context and retrieval: feeding the model real data~1h 25m
  21. 21fine-tune a small specialist~2h 10m
  22. 22debugging broken ai output~1h 1m
  23. 23production tradeoffs~57m
  24. 24path checkpoint~1h 40m
  25. 25agent harnesses~1h 27m
  26. 26ai image generation~1h
  27. 27ai video generation~50m
  28. 28programmatic design~46m
  29. 29harness engineering~2h 6m

how it works

Learn through interactive exercises in your browser. Start with the foundations, then work through practice for your role. Many later exercises ask you to read, edit, and run code; the introductory lessons help you get started without prior coding experience. You can skip a stuck exercise and continue reading.

where you end up

reads, debugs, and directs any ai-generated python. ships and maintains production agent harnesses. ships a CLI agent build, then continues into harnesses and production tooling.

training a team? assign this course to the people doing this role, then review their work together. plan your team’s training.

One lesson is free in the browser. Plus includes curriculum access, saved progress, sync, and completion certificates for $12/month or $99/year. Team training is $297 per seat each month and stays sales-led.

Start this course