// roadmap
DevOps Roadmap
The path has 9 stages, from your first Linux command to running systems in production. Select a topic on the map to see what it covers and the guides, simulators, labs and quizzes we have for it. Mark topics as you go. Your progress stays in this browser.
- 9 stages
- 47 topics
- 30 core topics
- 20-29 weeks
- 27 practice projects
- 209 links to our material
Security focused? Try the DevSecOps roadmap.
$ roadmap status
- done
- 0 of 40 topics, 0 percent
- in progress
- 0
- skipped
- 0
- next
- Linux basics
- last opened
- nothing yet
// map
The path, stage by stage
Select a topic for what it covers and our material for it. Arrow keys move between topics, Esc closes the panel.
- core
- optional
- done
- in progress
- skipped
stage 01 of 09, 2-3 weeks
Fundamentals
Build a solid foundation with essential programming and system administration skills.
// before you start
Basic computer literacy, Willingness to learn
// outcomes
- Command line proficiency
- Basic programming skills
- System understanding
// practice projects
Personal Development Environment
beginner, 1-2 days
Set up a complete development environment with Linux, Git, and basic automation scripts.
Linux · Git · Bash · VS Code
System Monitoring Dashboard
intermediate, 3-5 days
Write Bash scripts that monitor system resources and generate reports.
Bash · Cron · Python · HTML/CSS
Automated Backup Solution
intermediate, 2-3 days
Build an automated backup system with shell scripts and cron jobs.
Bash · Cron · Rsync · Git
// roles it leads to
Junior DevOps Engineer, System Administrator, Platform Engineer Intern
- Files and permissions
- Processes and services
- Package managers
- Logs
Linux runs most servers, containers and cloud machines. Learn to move around the file system, read and change permissions, manage processes and services, and find the logs. Every later stage assumes you are at home in a Linux shell.
// read
// practice
// check yourself
- Variables and loops
- Pipes and redirection
- Exit codes
- Cron jobs
Bash glues tools together. Write scripts with variables, loops and functions, chain commands with pipes, and schedule jobs with cron. Small scripts like these remove most repetitive work.
// read
- Introduction to BashGuide, 10 parts
- How to Parse Command Line Arguments in BashPost, 9 min
- Looping Through File Content Line by Line in BashPost, 6 min
- How to Redirect All Output to a File in BashPost, 9 min
- Introduction to Bash Scripting (free ebook) (opens in a new tab)External, github.com
- ShellCheck (opens in a new tab)External, shellcheck.net
// practice
- Scripts and CLIs
- JSON and HTTP APIs
- Error handling
- Tests
You do not need to be a software engineer, but you need to read and write code. Pick Python or Go and learn enough to write small tools: parse JSON, call an HTTP API, handle errors and add a few tests.
// read
// check yourself
- TCP/IP and ports
- DNS
- HTTP and TLS
- SSH
Many outages look like network problems at first. Learn how packets move with TCP/IP, how names resolve with DNS, and what happens in an HTTP request and a TLS handshake. Then you can tell a firewall rule from a stale DNS record.
// read
// practice
// check yourself
- Commits and branches
- Merges and conflicts
- Pull requests
- Rebase
Git tracks every change to code and infrastructure. Learn commits, branches and merges, fix a merge conflict, and work through pull requests. CI/CD and GitOps build on the same workflow.
// read
// practice
// check yourself
stage 02 of 09, 2-3 weeks
Infrastructure as Code
Learn to provision and manage infrastructure through code.
// before you start
Linux basics, Basic programming
// outcomes
- Infrastructure automation
- Configuration management
- Reproducible deployments
// practice projects
Multi-Environment Infrastructure
intermediate, 1-2 weeks
Create development, staging, and production environments with Terraform.
Terraform · AWS/Azure · Ansible · Git
Start with:
AWS VPC with TerraformLab, 120 minAutomated Web Application Stack
advanced, 2-3 weeks
Deploy a web application stack with load balancers, databases, and monitoring.
Terraform · Ansible · Nginx · PostgreSQL · Prometheus
Infrastructure Testing Pipeline
advanced, 1-2 weeks
Build a CI/CD pipeline that tests infrastructure changes before deployment.
Terraform · Terratest · GitHub Actions · Checkov
// roles it leads to
Infrastructure Engineer, Platform Engineer, Cloud Engineer, DevOps Engineer
- Providers and resources
- Plan and apply
- State
- Modules
Terraform describes servers, networks and DNS records as code, then plans and applies the changes. Learn providers and resources, how state tracks what exists, and how modules keep code reusable.
// read
// practice
// check yourself
- Inventories
- Playbooks
- Roles
- Idempotence
Ansible configures servers over SSH with YAML playbooks. Learn inventories, playbooks and roles, and why every task should be safe to run twice.
- Templates
- Stacks
- Change sets
CloudFormation is the AWS infrastructure as code service. Learn it if your team runs only on AWS or inherits CloudFormation stacks. The ideas map closely to Terraform.
- Desired state
- Drift
- YAML
- Repository layout
Configuration management keeps servers and apps in a known, desired state. Describe that state in files, keep the files in Git so every change is reviewed, and learn to spot drift when reality no longer matches.
- Validate and lint
- Plan checks in CI
- Policy as code
- Terratest
Infrastructure code breaks like any other code. Check it before it reaches production: validate and lint, review the plan in CI, enforce rules with policy as code, and run real tests with tools like Terratest.
stage 03 of 09, 3-4 weeks
Containerization & Orchestration
Learn container technologies and orchestration platforms.
// before you start
Linux fundamentals, Basic networking
// outcomes
- Container expertise
- Kubernetes proficiency
- Microservices understanding
// practice projects
Microservices E-commerce Platform
advanced, 2-3 weeks
Build and deploy a microservices application with Docker and Kubernetes.
Docker · Kubernetes · Helm · Istio · PostgreSQL · Redis
Container CI/CD Pipeline
intermediate, 1-2 weeks
Create a pipeline that builds, tests, and deploys containerized applications.
Docker · GitHub Actions · Kubernetes · Harbor Registry
Kubernetes Cluster Setup
advanced, 1-2 weeks
Set up a production-ready Kubernetes cluster with monitoring and logging.
Kubernetes · Prometheus · Grafana · ELK Stack · Ingress
// roles it leads to
Container Platform Engineer, Kubernetes Administrator, Site Reliability Engineer, Platform Architect
- Images and Dockerfiles
- Volumes
- Compose
- Multi-stage builds
A container packages an app with everything it needs to run. Learn to write a Dockerfile, build and run images, keep data in volumes and run several services together with Compose.
// read
// practice
// check yourself
- Bridge networks and ports
- Service discovery
- CNI plugins
- Network policies
Containers talk to each other over virtual networks. Learn bridge networks and published ports in Docker, then how Kubernetes gives every pod an IP address, routes Services and limits traffic with network policies.
// read
// practice
- Pods and Deployments
- Services and Ingress
- Scheduling
- Autoscaling
Kubernetes runs containers across many machines and keeps them running. Learn pods, Deployments and Services, how the scheduler places work, and how the Horizontal Pod Autoscaler adds pods under load.
// read
// practice
// check yourself
- Charts
- Values
- Releases and rollbacks
Helm packages Kubernetes manifests into charts with values you can change per environment. Learn to install, upgrade and roll back releases, and write a small chart of your own.
- Small base images
- Non-root users
- Image scanning
- No secrets in layers
A container is only as safe as its image. Use small base images, run as a non-root user, scan images for known CVEs and keep secrets out of image layers.
// read
// practice
// check yourself
stage 04 of 09, 2-3 weeks
CI/CD Pipelines
Build automated deployment pipelines for continuous delivery.
// before you start
Git version control, Basic containerization
// outcomes
- Automated deployments
- Testing integration
- Release management
// practice projects
Multi-Stage CI/CD Pipeline
intermediate, 1-2 weeks
Build a pipeline with testing, security scanning, and deployment.
GitHub Actions · Docker · SonarQube · Kubernetes · ArgoCD
GitOps Deployment System
advanced, 2-3 weeks
Set up a GitOps workflow for automated deployments with ArgoCD.
ArgoCD · Helm · Kubernetes · Git · Prometheus
Blue-Green Deployment Pipeline
advanced, 1-2 weeks
Create a zero-downtime deployment with automated rollback.
Jenkins · Kubernetes · Helm · Monitoring · Load Balancer
// roles it leads to
Release Engineer, DevOps Engineer, Platform Engineer, CI/CD Specialist
- Workflows and jobs
- Runners
- Secrets
- Caching
GitHub Actions runs your build, tests and deploys on every push. Learn workflows, jobs and steps, how runners work, and how to use secrets and caches safely.
// read
// practice
// check yourself
- Jenkinsfile pipelines
- Agents
- Plugins
Jenkins is a self-hosted CI server that many companies still run. Learn it if your team uses it: pipelines in a Jenkinsfile, build agents and plugins.
- .gitlab-ci.yml
- Stages and jobs
- Runners
GitLab CI is built into GitLab. Pipelines live in .gitlab-ci.yml as stages and jobs and run on GitLab runners. Most ideas carry over from GitHub Actions.
- GitOps principles
- Sync and drift
- Multi-environment repos
With GitOps, Git holds the desired state of your clusters and a controller makes each cluster match it. Argo CD is the most common controller. Learn sync and drift, and how to lay out a repo for several environments.
// read
// practice
// check yourself
- Tests in CI
- Preview environments
- Blue-green and canary
- Feature flags
A pipeline is only useful if it stops bad changes. Run tests on every change, give reviewers a preview environment, and ship with blue-green or canary releases that you can roll back.
// read
// practice
// check yourself
stage 05 of 09, 4-6 weeks
Cloud Platforms
Learn major cloud service providers and their services.
// before you start
Infrastructure as Code, Networking fundamentals
// outcomes
- Multi-cloud expertise
- Cost optimization
- Architecture design
// practice projects
Multi-Cloud Architecture
advanced, 3-4 weeks
Design and build an application that runs across AWS, Azure, and GCP.
AWS · Azure · GCP · Terraform · Kubernetes · Load Balancers
Serverless Application Suite
intermediate, 2-3 weeks
Build a serverless application with API Gateway, Lambda, and DynamoDB.
AWS Lambda · API Gateway · DynamoDB · CloudFormation · S3
Cost Optimization Dashboard
advanced, 1-2 weeks
Create automated cost monitoring and optimization recommendations.
CloudWatch · AWS Cost Explorer · Python · Grafana · Lambda
// roles it leads to
Cloud Engineer, Solutions Architect, Cloud Security Engineer, Principal Engineer
- Droplets
- Managed databases
- Spaces object storage
DigitalOcean is a smaller cloud with simple pricing, which makes it a good first place to run real servers. Create a Droplet, add a managed database and describe it all in Terraform.
- IAM
- EC2 and S3
- VPC networking
- Lambda
AWS is the largest cloud, so most job posts mention it. Start with IAM, EC2, S3 and VPC networking, then Lambda. Learn how the pieces connect before you learn every service.
- Resource groups
- Virtual machines
- AKS
Azure is common in companies that already use Microsoft tools. Learn resource groups, virtual machines, networking and AKS, its managed Kubernetes. The core ideas match AWS under different names.
- Projects and IAM
- Compute Engine
- GKE and Cloud Run
Google Cloud is strong in Kubernetes and data tools. Learn projects and IAM, Compute Engine, GKE and Cloud Run.
- Cost allocation tags
- Right-sizing
- Reserved and spot
- Data transfer
Cloud bills grow quietly. Tag resources so every cost has an owner, right-size machines, use reserved and spot capacity where it fits, and watch data transfer fees.
// read
// check yourself
- Well-Architected pillars
- High availability
- Exam practice
The AWS Certified Solutions Architect Associate exam checks that you can design systems on AWS. It is optional, but some employers filter on it. Build real things first, then study the Well-Architected pillars.
// read
// check yourself
stage 06 of 09, 2-3 weeks
Monitoring & Observability
Set up monitoring and observability for your systems.
// before you start
Container basics, Cloud fundamentals
// outcomes
- System visibility
- Incident response
- Performance optimization
// practice projects
Complete Observability Stack
advanced, 2-3 weeks
Build end-to-end monitoring for a microservices application.
Prometheus · Grafana · Jaeger · ELK Stack · AlertManager
SLO Monitoring Dashboard
advanced, 1-2 weeks
Track SLIs and SLOs with automated alerts and error budgets.
Prometheus · Grafana · SLO Library · PagerDuty · Kubernetes
Start with:
Complete Monitoring Stack SetupLab, 150 minPerformance Analysis Tool
intermediate, 1-2 weeks
Build automated detection of performance regressions.
Prometheus · Python · Grafana · Statistical Analysis · Alerts
// roles it leads to
Site Reliability Engineer, Observability Engineer, Platform Reliability Engineer, Principal SRE
- Metrics and exporters
- PromQL
- Dashboards
- Alerting
Prometheus scrapes metrics from your services and stores them as time series. Grafana turns them into dashboards. Learn PromQL, build a dashboard and write alerts that fire only for real problems.
// read
// practice
// check yourself
- Structured logs
- Log shipping
- Search and retention
Logs tell you what happened and why. Write structured logs, ship them from many machines to one place, and search them with Elasticsearch and Kibana or with Grafana Loki. Keep an eye on what they cost.
- Latency percentiles
- SLOs and error budgets
- On-call
- Incident response
Users feel latency and errors, not CPU graphs. Measure request latency at p95 and p99, set SLOs with error budgets, and use them to decide when to page someone. Then practice running an incident.
// read
// practice
// check yourself
- Spans and traces
- Context propagation
- OpenTelemetry
In a system of many services, one slow request can cross ten of them. Tracing follows the request with spans so you can see where the time went. OpenTelemetry is the standard way to add it.
// read
// practice
// check yourself
stage 07 of 09, 3-4 weeks
Security & Compliance
Build security checks into your pipelines and platforms.
// before you start
CI/CD pipelines, Container security basics
// outcomes
- Security automation
- Compliance management
- Threat mitigation
// practice projects
Secure CI/CD Pipeline
advanced, 2-3 weeks
Build a pipeline with security scanning and compliance checks.
SAST/DAST · Container Scanning · Secret Detection · Policy as Code
Zero Trust Network
advanced, 2-3 weeks
Apply zero-trust principles with a service mesh and mTLS.
Istio · Cert-Manager · OPA Gatekeeper · Falco · Network Policies
Compliance Automation
intermediate, 1-2 weeks
Automate compliance reports and fixes for SOC 2 or GDPR.
Open Policy Agent · Falco · Cloud Security Tools · Automation Scripts
// roles it leads to
DevSecOps Engineer, Security Engineer, Compliance Engineer, Security Architect
- SAST
- Dependency scanning
- Supply chain
- Security gates
Find problems before attackers do. Add static analysis, dependency checks and image scans to your pipeline, and decide which findings block a merge.
// read
- Static Application Security Testing (SAST)Guide, 4 parts
- Security GatesGuide, 4 parts
- OWASP Top 10Guide, 11 parts
- Dependency Scanning: Finding Vulnerabilities Before Attackers DoPost, 10 min
- How to Integrate DAST Into Your CI/CD Pipeline (With OWASP ZAP Examples)Post, 10 min
- Software Supply Chain Security: SBOMs, Sigstore, and SLSA in PracticePost, 11 min
- Vault
- SOPS
- Rotation
- Leaked secrets
Passwords and API keys do not belong in Git or in images. Keep them in a secrets manager like Vault, encrypt config files with SOPS, give each app only the secrets it needs and rotate them.
// read
// practice
// check yourself
- Least privilege
- Cloud IAM
- Kubernetes RBAC
- OAuth and OIDC
Every person and service should have only the access it needs. Learn cloud IAM policies, Kubernetes RBAC, OAuth and OIDC for logins, and how to lock down SSH.
// read
// practice
// check yourself
- Policy as code
- Audit evidence
- Threat modeling
Audits like SOC 2 ask you to prove that your controls work. Write the rules as code, check them in CI and collect the evidence automatically. Threat modeling helps you pick which controls matter.
- Pod Security Standards
- Capabilities and seccomp
- Runtime detection
Some attacks start after deploy. Apply Pod Security Standards, drop Linux capabilities, add seccomp profiles and watch for odd behaviour at runtime. Patch the kernel and the runtime when CVEs land.
// read
// practice
// check yourself
stage 08 of 09, 2-3 weeks
Database Management
Handle data persistence, scaling, and database operations.
// before you start
Basic programming, Cloud fundamentals
// outcomes
- Database expertise
- Data reliability
- Performance optimization
// practice projects
Database Migration Pipeline
advanced, 2-3 weeks
Automate a migration from a legacy system to a cloud database.
Database Migration Service · ETL Tools · Monitoring · Rollback Strategies
Multi-Database Architecture
advanced, 2-3 weeks
Design polyglot persistence with different database types.
PostgreSQL · MongoDB · Redis · Elasticsearch · Data Sync
Database Monitoring System
intermediate, 1-2 weeks
Build performance monitoring for your databases.
Prometheus · Grafana · Database Exporters · Alert Rules
// roles it leads to
Database Engineer, Data Platform Engineer, Database Administrator, Data Architect
- SQL queries
- PostgreSQL
- Redis and MongoDB
- Choosing a database
Most apps keep their data in PostgreSQL or MySQL, with Redis or MongoDB for some jobs. Learn SQL, how tables and indexes work, and when a document or key-value store is a better fit.
// read
// practice
// check yourself
- Migrations as code
- Schema changes in CI
- Zero-downtime changes
Schema changes should go through the same review and pipeline as code. Use a migration tool, run migrations in CI against real data, and change a busy table without locking it.
// read
// check yourself
- Backup types
- Point-in-time recovery
- Restore drills
- RPO and RTO
A backup you have never restored is a guess. Learn full and incremental backups, point-in-time recovery and restore drills, and agree on how much data and time the business can afford to lose.
- EXPLAIN and query plans
- Indexes
- Connection pooling
- Caching
Slow queries are a common cause of slow apps. Read query plans with EXPLAIN, add the right indexes, pool connections and cache hot reads.
// read
// practice
- Replication
- Verification
- Cutover and rollback
Moving data between servers, versions or vendors needs a plan. Copy it with replication, check that both sides match, and cut over with a short, tested way back.
stage 09 of 09, ongoing
Continuous Learning
Keep learning and share what you know.
ongoing habits, no progress to track
// before you start
Curiosity, Growth mindset
// outcomes
- Continuous growth
- Community impact
- Knowledge sharing
// practice projects
Open Source Contribution
advanced, ongoing
Contribute to a DevOps tool you use, or start your own project.
GitHub · Community Building · Documentation · Code Review
Technical Blog Series
intermediate, ongoing
Share your DevOps work and what you learn in regular blog posts.
Writing · SEO · Community Engagement
Mentorship Program
intermediate, ongoing
Guide junior engineers and help the community grow.
Leadership · Communication · Knowledge Transfer · Career Coaching
// roles it leads to
Senior Engineer, Staff Engineer, Principal Engineer, Distinguished Engineer
- Read changelogs
- Break things on purpose
- A daily incident
Tools change every year, but the habit of trying them carries over. Read changelogs, keep a small lab to break things in, and when something fails, find out why.
- First contributions
- Reading other code
- Good pull requests
Contributing to open source teaches you to read large codebases and to work with strangers through pull requests. Start with docs and small fixes in tools you already use.
- Runbooks
- Postmortems
- Blog posts
Clear writing spreads what you know. Write runbooks, design docs and blameless postmortems, and turn what you learn into a blog post.
- Pairing
- Code review
- Sharing notes
Teaching others shows you the gaps in what you know. Pair with junior engineers, give kind and specific code reviews, and share your notes.
- Ship something small
- Run it yourself
- Write it up
A small project you run yourself teaches more than a course. Deploy something real, give it monitoring and backups, and write down what broke.
- Recorded talks
- Workshops
- Local meetups
Talks and workshops show how other teams solve the same problems. Many conferences post their talks online for free, so you can learn without travelling.
- Forums and chat
- Meetups
- Newsletters
You learn faster with people around you. Join a forum, a chat group or a local meetup, ask questions, and answer a few.
// career notes
Career notes
US pay and job growth for the jobs this roadmap leads to. The BLS has no DevOps job title, so each row is the closest occupation it tracks. Pay in other countries can be very different.
| Occupation (BLS) | Roles on this roadmap | Pay, 10th to 90th percentile | Median | Jobs, 2025-2035 |
|---|---|---|---|---|
| Network and computer systems administrators (BLS page, opens in a new tab) | System administrator, junior DevOps engineerFundamentals, IaC | $62,640 to $155,050 | $99,130 | -4% |
| Software developers (BLS page, opens in a new tab) | DevOps, platform and release engineer, SREContainers, CI/CD, Monitoring | $82,460 to $214,670 | $135,980 | +10%Growth covers software developers, QA analysts and testers together. |
| Computer network architects (BLS page, opens in a new tab) | Cloud and network architectCloud | $79,900 to $202,680 | $134,050 | +8% |
| Information security analysts (BLS page, opens in a new tab) | DevSecOps and security engineerSecurity | $75,090 to $199,850 | $129,180 | +21% |
| Database administrators (BLS page, opens in a new tab) | Database administratorDatabases | $60,230 to $163,320 | $104,620 | 0% |
| Database architects (BLS page, opens in a new tab) | Database and data platform engineerDatabases | $86,240 to $204,000 | $139,500 | +9% |
Source: US Bureau of Labor Statistics, Occupational Outlook Handbook. Pay for May 2025, job projections for 2025 to 2035, pages updated 27 August 2026. We checked the numbers on 7 November 2026.
// getting the job
Practice the questions interviewers ask, and read how the role is changing.
Browse interview questionsFound a wrong link or a missing topic? Open an issue on GitHub.