DevOps Engineering study plan
DevOps Engineering · Give each week a clear focus.
8 weeks · About 7 hours per week
Continuous Integration and Continuous Delivery (CI/CD) Pipelines
Map the anatomy of a CI/CD pipeline stage by stage, then write a one-page reference covering build triggers, artifact promotion, and environment gating
Configure a pipeline in a self-hosted runner that runs unit tests and publishes a versioned artifact on every merge to main
Implement branch protection plus required status checks so failing builds block merges
Diagnose three common pipeline failure modes (flaky tests, cache poisoning, missing secrets) and document the fix for each
Your checkpoint
A working repository with a test-and-package pipeline, branch protection enabled, and a README recording the three failure-mode diagnoses and fixes
Continuous Integration and Continuous Delivery (CI/CD) Pipelines
Add a deployment stage that promotes the same immutable artifact across staging and production with manual approval gates
Implement automated rollback triggered by a failing post-deploy health check
Introduce build caching and parallel test sharding, then measure and record the pipeline duration before and after
Write a deployment strategy comparison covering blue-green, canary, and rolling updates with the trade-offs that decide between them
Your checkpoint
A pipeline that deploys one artifact through two environments with an approval gate and automatic rollback, plus a before/after duration table and strategy comparison note
Infrastructure as Code and Configuration Management
Provision a small environment (network, compute, storage) declaratively with a state-managed IaC tool and commit the configuration
Refactor duplicated configuration into reusable modules or roles with typed input variables
Configure remote state with locking and demonstrate that two concurrent applies cannot corrupt it
Run a plan/dry-run diff and explain every proposed change before applying
Your checkpoint
A modular IaC repository that creates and destroys a full environment from scratch, with remote locked state and a written plan review of one apply
Infrastructure as Code and Configuration Management & Scripting, Automation, and Tooling
Write an idempotent configuration-management playbook that converges a server to a defined state and produces no changes on a second run
Add drift detection that reports when live infrastructure diverges from declared state
Script a repeatable environment teardown and rebuild using a single command with no manual steps
Automate linting and format checks for IaC and scripts in the pipeline so style violations fail the build
Your checkpoint
A one-command environment rebuild plus a drift-detection report showing a clean second run and a deliberately introduced drift being caught
Cloud Platforms, Containers, and Orchestration
Containerize an application with a minimal, multi-stage image and pin base image digests
Deploy that image to a local or managed orchestrator with health probes, resource requests and limits, and a horizontal scaling rule
Configure a rolling update and a deliberate failed rollout, then observe how the orchestrator recovers
Explain how service discovery, ingress, and persistent storage are wired in the cluster you built
Your checkpoint
A running orchestrated deployment with probes, limits, autoscaling, and a written postmortem of the failed rollout showing recovery behavior
Monitoring, Observability, and Site Reliability Engineering
Instrument an application with structured logs, metrics, and distributed traces that share a correlation identifier
Define service-level indicators and objectives for one user-facing journey and derive an error budget
Build a dashboard and two alerts: one symptom-based burn-rate alert and one saturation alert, each with a runbook link
Write a blameless postmortem for a simulated incident using the telemetry you collected
Your checkpoint
A dashboard, two tested alerts with runbooks, and a postmortem that cites specific log lines, metric panels, and trace spans
DevSecOps, Security, and Compliance Automation & Version Control, Branching, and Release Engineering
Scan dependencies, container images, and IaC for vulnerabilities in the pipeline and fail builds on a defined severity threshold
Add secret scanning and a pre-commit hook that blocks credentials before they reach the remote
Choose a branching model for a two-week release cadence and document the merge, hotfix, and tagging rules
Automate changelog generation and semantic version tagging from commit history
Your checkpoint
A pipeline with security gates and secret scanning enabled, plus a written branching and release policy with an auto-generated changelog and version tag
Scripting, Automation, and Tooling & Collaboration, Agile Practices, and DevOps Culture
Automate one recurring toil task with a tested script that has error handling, logging, and a dry-run flag
Run a full review pass over all eight domains, listing weak areas and scheduling targeted practice for each
Draft a team working agreement covering code review turnaround, incident communication, and blameless retro practices
Complete a timed mixed-domain practice set and analyze every miss by root cause
Your checkpoint
A toil-automation script with tests, a personal weak-area remediation list, and a one-page team working agreement
Milestones
Week 2: You can take one commit from merge to a gated production deploy and back out automatically without touching a console
Week 4: You can destroy and rebuild an entire environment from code alone, and a second configuration run reports zero changes
Week 6: Given a failing service, you can locate the cause using logs, metrics, and traces within ten minutes and name the breached SLO
Week 7: A deliberately vulnerable dependency or leaked secret in a test commit is blocked by the pipeline before merge
Week 8: You score above 80 percent on a timed mixed-domain set and have a written plan for every remaining weak area