AI Development v6.2 · Claude + Codex
One does the work, the other checks
Generating is easy now; the work is checking. You’ll put together your own pilot with a briefing, a critiqued plan, a review of what changed, tests, and acceptance, choosing the model and effort for each step.
start now →6 modules · 30 lessons of ~15 minutes · at your own pace
What you’ll be able to do
Criteria before codea briefing with a goal, what’s in and out of scope, acceptance criteria, and limits you can check.
Cross-review that countsone model does the work, the other examines the actual work, and the tests and you decide.
Costs under controlmodel and effort chosen for each step, with time and usage measured in your pilot.
Who it’s for
This is for you if you
- Already use Claude Code or Codex and accept the result because it looks right.
- Deliver code, automation, or reports made with AI to other people.
- Prefer to learn by doing.
This isn’t for you if you
- Want to switch from one assistant to the other: see the Claude → Codex course.
- Want traditional video lessons.
