Delivery Playbook · Roles
Engineering roles
Solution Architect / Tech Lead, Software Engineers and AI / Data Engineer.
Engineering roles design, build and own the software. The Tech Lead is accountable for technical quality; every engineer is responsible for the changes they commit, whether they wrote them or an AI tool helped.
Solution Architect / Tech Lead
TechSoftOwns technical direction, architecture and engineering quality.
Owns
- Architecture and integration design
- Technical decisions and records (ADRs)
- Code review standards and technical risk
- Non-functional requirements: performance, security, scalability
Day to day
- Assesses existing systems before work starts
- Breaks work into increments that can be released safely
- Reviews designs and critical code
- Coaches engineers, including on effective AI-assisted practice
Use of AI: Sets the rules for where AI-written code is acceptable and how it is reviewed. Responsible for the result.
Works with: Works with Business Analyst on scope, with Security and DevOps on controls.
Software Engineers (backend, frontend, mobile)
TechSoftBuild, test and ship working software in small increments.
Owns
- Implementation quality of their changes
- Unit and integration tests
- Technical documentation of what they build
Day to day
- Implement stories to the definition of done
- Review each other's changes
- Fix defects and pay down technical debt
- Support releases and production issues
Use of AI: Use AI coding tools to accelerate work; understand, test and take responsibility for every line they commit.
Works with: Hand finished increments to QA and the release process.
AI / Data Engineer
TechSoftBuild the data and AI components that products and workflows depend on.
Owns
- Data integration and pipelines
- AI agent and workflow design, prompts and evaluations
- Monitoring of AI behavior in production
Day to day
- Chooses AI versus deterministic automation per step
- Builds permissions, approvals and audit trails
- Evaluates quality against agreed test sets
- Tunes and monitors after launch
Use of AI: Responsible for how the AI features delivered to the customer behave.
Works with: Works with Business Analyst, Security and the customer's process owners.
| Discover | Design | Build | Verify | Operate | |
|---|---|---|---|---|---|
| Tech Lead / Architect | •• | ••• | ••• | •• | •• |
| Software Engineers | • | ••• | •• | •• | |
| AI / Data Engineer | • | •• | ••• | •• | •• |
Engineering habits
- Small, reviewable changes: easy to understand and easy to roll back
- Architecture decisions recorded as short ADRs
- Tests written with the change, not afterwards
- Documentation updated in the same change as the code
- No change is merged that the author cannot explain
Related
Lifecycle
Stage 3: Build
Deliver working software in small, reviewed steps, using AI to help and with people responsible for the result.
Quality & AI
AI-assisted engineering
Where we use AI, the rules that keep it safe, and who is responsible.
Quality & AI
Quality gates
Four gates a change passes between idea and production, and what each one checks.