· 5 min read
A single AI course for the whole organisation tends to be too shallow for engineers and too technical for everyone else. A role-based framework avoids this by defining what each group needs to know and be able to do.
Leaders: decide, fund and govern
Leaders need enough understanding to ask good questions, weigh value against risk, and sponsor change. Their learning focuses on capabilities and limits, investment choices, accountability and how to support teams adopting AI.
Practitioners: apply AI in daily work
Practitioners in business and delivery functions need applied skills tied to their own tasks, plus clear rules for data handling and review. Examples should come from their workflows, not generic demonstrations.
Builders: design, evaluate and operate
Builders create AI-enabled systems. They need depth in application design, retrieval, tool use, evaluation, observability and release practice, and they benefit from labs using their own stack.
Connect the levels
- Use a shared vocabulary across all three levels.
- Define what each level can approve, what it must escalate and to whom.
- Plan reinforcement, such as champions or communities of practice, so skills are used after the sessions end.