The Story
Follow the narrative to build a mental model of how everything connects.
The Alignment
Before a single pipeline runs, teams must agree on how they work together. Agile ceremonies set the cadence, value stream mapping exposes where work stalls, and SRE thinking introduces error budgets so reliability and velocity stop fighting each other. DevOps starts as a culture problem, not a tooling problem.
The Backlog
Every unit of work lives in Azure Boards — epics break into features, features into stories, stories into tasks. Kanban boards make flow visible; a WIP limit stops a team from starting ten things and finishing none. Traceability threads a single work item all the way through to the commit, the build, and the release that shipped it.
The Fork in the Road
Every team must choose how code branches. Trunk-based development keeps branches short and merges constantly — fast, but demands strong automated testing and feature flags to hide unfinished work. GitFlow keeps long-lived release branches — safer for scheduled releases, slower to integrate. Release flow, Microsoft’s own model, branches from main per release and only ever fixes forward.
The Gatekeeper
A branch is only as trustworthy as the policy protecting it. Required reviewers, linked work items, and passing builds all have to clock in before a pull request can merge. CODEOWNERS automatically pulls in the right expert for the right file. Forks keep outside contributors at arm’s length from anything protected.
The Assembly Line
A push lands, and the build pipeline wakes up. YAML defines every stage as code — build, unit test, package — versioned right alongside the source it compiles. Multi-stage pipelines chain the whole journey from commit to artifact in a single file, triggered by CI pushes, pull requests, schedules, or another pipeline’s own output.
The Warehouse
Nothing gets built from scratch every time. Azure Artifacts feeds host NuGet, npm, Maven, Python, and Universal Packages, with upstream sources proxying the public internet so a single internal feed becomes the source of truth. Semantic versioning is the contract that stops today’s build from silently breaking tomorrow’s consumer.
The Launch Sequence
A build artifact becomes a release only after passing through staged environments — dev, test, staging, production — each one a gate for the next. Approvals pause the pipeline for a human sign-off; automated checks pause it for a machine one. This is where 50%+ of the exam lives: the mechanics of getting code safely, repeatably, into production.
The Blast Radius
Shipping to everyone at once is a bet you don’t need to make. Blue-green keeps two identical environments and flips traffic between them instantly — rollback is just flipping back. Canary sends a sliver of real traffic to the new version and watches. Ring-based deployment goes further, expanding in concentric circles from an internal ring outward to the whole customer base.
The Blueprint
Infrastructure is code now, not a ticket to the platform team. Bicep compiles down to ARM and deploys idempotently — run it twice, get the same result. Terraform trades that Azure-native tightness for multi-cloud reach, tracking reality in a state file and previewing every change with a plan before it touches anything.
The Inspection
Security that only happens right before release is security that happens too late. SAST reads the source without running it, catching injection flaws and hard-coded secrets at the PR stage. DAST attacks a running instance the way an actual adversary would. SCA checks whether any of the hundred open-source packages you depend on have a known CVE attached.
The Vault
A secret typed into a pipeline’s YAML is a secret that is no longer secret. Azure Key Vault holds it instead, and a variable group links the pipeline to the vault without ever exposing the value in source control. Service connections increasingly skip stored credentials entirely, authenticating via managed identity or workload identity federation.
The Feedback Loop
The pipeline doesn’t end at deployment — it ends when production tells you the truth. Application Insights streams telemetry back the moment users touch the new release. Azure Monitor alerts can halt or roll back a bad deployment automatically. DORA metrics turn all of it into a scoreboard: how often you ship, how fast you recover, how often it breaks.
