High-velocity IT techniques and practices
What this domain tests
The exam checks that you can select and justify HVIT techniques - Lean, Agile, DevOps, site reliability engineering (SRE) and supporting ITIL practices - and explain how each improves speed or feedback.
Core concepts
- Agile provides empirical control: transparency, inspection, adaptation, delivered in short increments with a working product as the primary measure.
- DevOps extends this across the lifecycle with three concerns: flow (small batches, removed bottlenecks), feedback (fast detection of problems), and continual learning (experimentation, blameless review). Common enablers: continuous integration and continuous delivery (CI/CD), automated testing, infrastructure as code, feature toggles.
- SRE manages reliability as an engineering problem with service level indicators, service level objectives and error budgets that trade reliability against change rate explicitly.
- Techniques the syllabus expects you to know: value stream mapping, Kanban and work-in-progress limits, minimum viable product (MVP), hypothesis-driven development, A/B testing, chaos engineering, blameless post-mortem.
- Supporting ITIL practices: deployment management (how changes reach the live environment), release management (how features are made available to users), service validation and testing, problem management (known errors), continual improvement.
Worked example
A team wants a recommendation engine. Rather than building the full feature, it states a hypothesis: "personalised recommendations raise click-through by 5% within four weeks". It builds the smallest version that can test this, hides it behind a feature toggle, runs an A/B test, and compares cohorts. If the hypothesis holds, it scales; if not, it pivots and keeps the learning. Reasoning: the MVP exists to buy information cheaply, and the toggle allows instant rollback.
Traps that cost marks
- Treating DevOps as a toolchain purchase rather than flow, feedback and learning.
- Assuming Agile means no planning or documentation; it means short plans and just-enough documentation.
- Defining MVP as "half a product" instead of the smallest thing that tests the hypothesis.
- Automating a wasteful process - the guiding principle is optimize and then automate.