Module 6: Teaching and Leading AI-Augmented Teams
Learning objectives
By the end of this module, you will be able to:
- Structure a ten-minute explanation of a technical concept using the problem-before-solution pattern, and deliver it without slides.
- Pick the right level of abstraction for an audience and adjust when you see you got it wrong.
- Run a three-phase adoption plan that moves a team from individual experiments to a shared workflow with a rules file, review norms and an evaluation gate.
- Choose metrics that measure delivery outcomes rather than tool usage, and explain why "acceptance rate" is the wrong number.
- Recognise the three predictable objections to AI-assisted development and answer each one honestly.
Module content
Chapter 1: Teaching a technical concept
A repeatable structure for explaining anything technical in ten minutes: the problem, the naive approach and why it fails, the concept, one worked example, one thing that goes wrong. With a full worked example (explaining RAG to a mixed audience).
Chapter 2: Adoption and measurement
Three phases - explore, standardise, scale - with the artefacts each one produces. What to measure (DORA metrics plus three AI-specific signals), what not to measure, and how to talk about the results.
Quiz
Six questions.
Why a course on engineering ends with teaching
Because in an AI-augmented team, the scarce skill is not writing code; it is transferring judgement. The engineer who can explain why the test comes before the generation, why the spec names non-goals, why the agent is read-only by default, multiplies the practice across a team. The one who can only do it multiplies nothing.