All chapters
  1. Start hereEnterprise AI adoption strategy, stage by stage and role by role
  2. Chapter 1Why AI adoption stalls
  3. Chapter 2Governance is the first enabler
  4. Chapter 3Who decides, who runs it
  5. Chapter 4Finding the right use cases
  6. Chapter 5The adoption ladder
  7. Chapter 6Role tracks
  8. Chapter 7Reuse first, as code
  9. Chapter 8The worked example
  10. Chapter 9The enablement program
  11. Chapter 10Measuring adoption honestly
  12. Chapter 11Evaluating and mapping tools
  13. Chapter 12Hands-on labs
  14. Chapter 13Templates, each filled
  15. Chapter 14Glossary
  16. AboutAbout this site
Chapter 6

Role tracks

Nineteen roles, one template: what each role decides, how it adopts AI stage by stage, the evidence it owns and who it hands off to.

No organization adopts AI as one crowd, and a large regulated one least of all. A chief information officer (CIO) adopts it through funding and risk appetite. A developer adopts it through a coding agent and then by building AI features into products. A production support engineer adopts it at three in the morning, during an incident. One common ladder cannot describe all of that, so this chapter gives nineteen role tracks, each written to the same template. Every track opens with your first steps and ends that section with the fully adopted workflow: how the work runs at Stage 4 to 5 and what changes for the organization.

Treat each track as a starting shape: rename the roles to match your org chart and tune the steps to your own workflow.

1. Start from what you need to do

I need to…Start with
Fund and steer AI adoption, set risk appetiteExecutive sponsor
Lead a team through the change, protect reviewer timeEngineering manager · Project and delivery lead
Shape what gets built and accept itProduct owner · Business analyst · Experience designer
Build software and agentic productsArchitect · Developer · AI engineer
Prepare data and the platformData engineer and steward · Platform engineer and FinOps
Test and produce first-line release evidenceQA engineer
Independently challenge or validateIndependent validation
Release and run in productionRelease manager · Production support and SRE
Govern and secureRisk and compliance · Security
Use AI in customer operationsBusiness operations
Run the enablement programAI enablement lead

2. From requirement to production, every role on one path

AI does not wait for full adoption to reach production. At Stage 1 every change still reaches production through the normal process; AI only helps people draft. What adoption changes is how much of each step agents do, and how much evidence surrounds it. The accountable decision at each step stays with the same role.

StepRolesAt the start (Stage 1 to 2)Fully adopted (Stage 4 to 5)Decision that stays human
AskExecutive sponsor · Product ownerSummarize the request with an approved tool; log the use case at intakeAgents turn an accepted intake into a brief, a risk-tier draft and a metricFund it, set the risk tier, accept the outcome
ShapeBusiness analyst · Experience designer · ArchitectDraft stories and criteria; a human rewrites each oneImpact analysis across every service; the reuse check runs before any designReuse, extend or create; accept the stories
BuildDeveloper · AI engineer · Data engineer · Platform engineerExplain code, suggest a test; the human types the changeA bounded workflow drafts the pull request in a sandbox and fixes failing checksApprove the merge (never the requester)
TestQA engineer · Independent validationSuggest edge cases; QA choosesGenerated tests and evaluation sets on every change; independent validation for high-tier usePass or fail against thresholds; accept residual risk
ReleaseRelease managerDraft release notesThe change record assembles itself from CI, tests and approvalsGo or no-go; rollback
RunProduction support · Business operationsSummarize logs and ticketsAgents correlate alerts and propose runbook steps; staff use AI with customers behind a fallback pathIncident command; every action on production
Govern, throughoutRisk and compliance · Security · AI enablement leadApprove tools, set acceptable use, run the first labsMonitoring, periodic review and material-change review run on evidence the pipeline already producesSet the standards; challenge; stop the work

3. In any organization

The tracks are written for a large regulated organization because that is the hardest case. A smaller or unregulated organization follows the same order with fewer people and lighter paperwork.

In a large regulated organizationIn a small or unregulated organization
Nineteen separate rolesOne person covers several tracks: the tech lead is architect and release manager; a founder is the executive sponsor
A second-line risk function and independent validationA peer from another team challenges high-impact use cases
Formal change authority and change recordsA pull request template and a short release checklist
Full evidence retention under policyKeep the traces, decision records and evaluation results you would need to explain an incident

What never scales down: no customer or secret data in unapproved tools; the requester of an agent task never approves it; a kill switch for anything that acts on its own; a human in command of incidents.

4. Two kinds of adoption, on every track

KindWhat it meansExample
AI in how you workThe role's own tasks: drafting, analysis, review, triageA business analyst drafts stories from approved policy text
AI in what you deliverThe product, platform, control or service the role is responsible for contains AIAn AI engineer ships a policy assistant; security threat-models it; operations staff use it with customers on the line

The ladder stage describes the controls, context and evaluation around a use case. It is not seniority, and it is never permission to hand an accountable decision to an agent. Every track names at least one activity that stays human-operated even at Stage 5.

5. Decision relationships

"Who approves" is too coarse once AI is in the work. Every track uses five relationships.

RelationshipMeansExample
Owner or approverHolds the decision and its accountabilityThe product owner accepts the feature
Independent validator or challengerTests the claim without being part of the team that built itIndependent validation reviews the policy assistant's evaluation design
Consulted control partnerGives a control view before the decisionPrivacy reviews the data flow
Evidence recipientReceives records to keep or assessInternal audit receives the inventory and validation evidence
NotifiedTold after the decisionThe contact-center lead learns of the release date

6. Nineteen tracks in eight families

Families follow the lifecycle. The badge on each track shows its line of defense: 1st line delivers and owns the risk, 2nd line sets standards and challenges, and internal audit (the 3rd line) assesses independently.

Lead

Shape

Build

Data & platform

Assure

Release & run

Govern

Adopt

7. The common template

Every track uses the same fifteen sections in the same order, so a reader can compare roles side by side.

#SectionWhat it answers
1At a glanceMandate, main decision, first safe win, biggest failure mode
2Your journeyFirst steps, the fully adopted workflow, what changes for the organization, how it looks in any organization
3Your decisionsWhat you own, advise or challenge, and what you never delegate
4Two kinds of adoptionAI in how you work, AI in what you deliver
5Your ladderStage 0 to 5: AI use, guardrail, evidence, decision relationship, step-down trigger
6Priority scenariosValue, risk tier, stage, first step, when not to use AI
7The due-date ask from your seatYour part of the worked example
8Hand-offsWhat you give, receive, challenge and hear about
9Evidence and recordsWhat you produce or consume, and where it is kept
10Skills and learning pathFoundation, practitioner, advanced, with labs
11Measures and incentivesHow the role is judged, with anti-gaming notes
12Cost, capacity and workforceBudget, reviewer time, jobs and skills
13Anti-patterns and stop conditionsWhat goes wrong and what stops the work
14Action kit and 30/60/90A copy-ready card and the first ninety days
15One diagramThe role's decision, evidence or operating loop

8. Separation of duties, in one table

ArtifactRequested or produced byApproved by
Epic briefPO, with an agent draftingBusiness owner
User storiesBA, with an agent draftingPO
Temporal-domain contractArchitectArchitecture review board, not the author
PRDeveloper or coding agentA different developer, under branch protection
Generated test suiteQA engineer or agentA different QA engineer or the test lead
Change recordRelease manager or agentChange authority
Validation reportIndependent validationUse-case owner or the policy-named authority accepts or approves the use, subject to recorded restrictions and residual risk

An agent is never an approver. When an agent does the work, the person who delegated it is the requester, and the approval goes to someone else with the existing authority.

9. Internal audit: the third-line interface

Internal audit is not a track and not part of the second line. It receives evidence (the AI inventory, control design, validation reports, change records, monitoring and exceptions) and independently assesses whether governance works. It never approves a first-line release and never designs controls for management. The risk and compliance and executive sponsor tracks show the interface in detail.