05

Engagement format

Practical AI Training and Mentoring for Teams

Work on the team’s real workflow and repository to produce a specification, working artifact, quality tests, and a transferable handoff.

Describe the task
01

Problem framing

What must be resolved

A team knows the tool names but has not practised turning its own workflow into a specification, artifact, test, and safe improvement cycle.

02

Fit

Suitable for

For teams and individuals learning RAG, agent workflows, Legal AI, and AI-native development.

03

Deliverables

What remains after the engagement

  1. Work plan
  2. Repository or prototype
  3. Tests
  4. Next-experiment list
04

Process

How the work proceeds

  1. 01Level diagnostic
  2. 02Task
  3. 03Iterative work
  4. 04Review
05

Scope

From input to a decision

When it is needed

Training and mentoring are appropriate when a team wants to transfer a way of working, not merely learn a tool. The material is a real workflow or repository in which a small, inspectable artifact can be built and assessed together.

Input material

Inputs include the team workflow or repository, starting level, intended artifact, and handoff owner. This keeps the work from becoming a demonstration of generic prompts detached from the next stage of practice.

What we do

The team creates a specification, bounded artifact, quality tests, and a decision record. The work separates implementation from verification, makes source boundaries visible, and shows how to return to a test after a code, data, or tool change.

Artifact and outcome

The result is a specification, working artifact, test set, and handoff runbook. Participants receive not a promise of competence but a concrete package they can extend, challenge, or safely retire.

Acceptance criterion

Acceptance occurs when the team can run the artifact, identify its quality check, explain the material boundary, and name the next-step owner without depending on the facilitator.

Boundary of responsibility

This is practical work on the team’s own material, not certification, an employment guarantee, or a replacement for internal ownership. If no workflow or handoff owner exists, the collaboration conditions should be prepared first.

Decision record

After a workshop, a team should be able to name its own artifact, verification criterion, source boundary, and person responsible for further work. A handover record captures those elements and open questions. It does not replace an internal tool decision or continuing organisational support.

06

Boundaries

Risks checked early

  • Topic too abstract
  • No practice time
  • No test data
07

Evidence path

Existing material that demonstrates the scope of work

08

FAQ

Practical questions

Can we work on our own project?

Yes, when context and anonymised examples can be shared safely.

Does training end with a tool presentation?

No. The work uses one real team workflow or repository and leaves a specification, artifact, test, and runbook. If that material is absent, preparation conditions are identified first rather than simulating an implementation.

How can a team tell it can take over the result?

The team should be able to run the artifact, identify its quality check, explain source boundaries, and name the next-step owner. That is a handoff criterion, not an individual competence certificate.

09

Scope and cost

Evidence before an estimate

I do not publish a fictional ‘from’. Scope and estimate follow a review of the workflow, material access, risk, expected artifact, and acceptance criterion.

Start by describing the workflow

The first response will identify whether this format fits and what is needed for an honest scope.

Start a conversation