Skip to content

Intermediate Track Preview

From Doer to Delegator

The beginner track is about becoming a Doer: someone who uses AI to build real software through conversation. You learned how to write effective prompts, work in a development environment, give AI persistent context, and deploy a live application.

The intermediate track picks up from there. It is about becoming a Delegator: someone who defines work clearly enough to hand entire tasks off to AI and be confident the results come back correct.

What does that shift look like in practice?

  • Decomposition: Breaking big goals into independently shippable pieces, each with its own user story and acceptance criteria, so you can delegate them one at a time
  • Skills: Encoding your team's repeatable processes into reusable instruction files that AI follows consistently, instead of re-explaining your conventions every conversation
  • Automated testing: Turning the acceptance criteria you already write into automated tests that verify your app works after every change, not just when you remember to check
  • Parallel delegation: Running multiple AI tasks at once, trusting your system of skills, tests, and context to catch problems without you watching every step

The shift is subtle but powerful. As a Doer, you prompt AI and review every response. As a Delegator, you define what "done" looks like, hand off the work, and verify the output against your criteria. You are managing AI the way a good manager manages people: clear expectations, clear contracts, clear verification.

What's in This Preview

These pages introduce two intermediate track concepts you can start using right now:

  1. Skills: Encoding Your Judgment: How to capture your team's processes as reusable AI instructions, so AI does things your way every time
  2. Automated Testing: Your Criteria Are Already Tests: How the Given/When/Then acceptance criteria you already write can become automated tests that catch problems before users do

These are self-contained previews, not the full intermediate lessons. They give you enough to try both techniques on whatever you are building. If you want the complete experience (decomposition, manual review, deployment pipelines, parallel workstreams), that is what the full intermediate track is for.