Book companion

Build · Ship · Operate

The DevOps Engineer Roadmap

Choose a DevOps direction, repair the gaps that matter, and turn engineering work into evidence another person can inspect.

Paths
3
Projects
3
Planning horizon
12 weeks
The DevOps Engineer Roadmap book cover: Build, Ship, Operate

The operating idea

DevOps is a set of responsibilities, not a collection of tools.

The roadmap starts with the work you want to own. Tools and certifications enter only when they solve a visible gap in delivery, infrastructure, or reliability.

Three primary paths

Choose the operating responsibility first.

Path A

Delivery & Cloud-Native Applications

You want to turn commits into dependable releases and own the application delivery path.

  • CI/CD and release controls
  • Containers and Kubernetes workloads
  • GitOps and deployment evidence
Proof to build

A versioned application that moves through a tested delivery pipeline into a documented environment.

Review this path
Path B

Infrastructure & Kubernetes Operations

You want to provision, operate, and recover the systems on which teams depend.

  • Linux and systems administration
  • Terraform and reproducible infrastructure
  • Kubernetes cluster operations
Proof to build

A reproducible environment with operating notes, failure tests, and a recovery record.

Review this path
Path C

Reliability & Observability

You want to understand system behavior, define useful signals, and improve recovery.

  • Metrics, logs, traces, and SLOs
  • Incident analysis and alert quality
  • Failure drills and recovery evidence
Proof to build

An observable service with explicit reliability targets, a failure drill, and an evidence-based improvement.

Review this path

Portfolio sequence

Build, operate, and recover.

Each project adds an operating responsibility and leaves a record of decisions, failures, and limitations.

  1. 1
    From Commit to Deployment

    Build a small service and a pipeline that tests, packages, deploys, and records a release.

    Open the workspace
  2. 2
    Reproducible Infrastructure and Kubernetes

    Provision a bounded environment, deploy a workload, and document how another person can reproduce and recover it.

    Open the workspace
  3. 3
    Observe, Break, Recover

    Instrument a service, define useful signals, inject one controlled failure, and improve the recovery path.

    Open the workspace

Reader resources

Continue from the book without losing its context.