Prerequisites: Linux & Networking

DevOps runs on Linux. Learn: file system navigation, permissions, process management, systemd, cron jobs, shell scripting (bash). Networking: TCP/IP, DNS, HTTP/HTTPS, firewalls, load balancing, SSH. You don't need to be a sysadmin, but you must be comfortable in a terminal. Spend 2-3 weeks on Linux commands before touching any DevOps tools. Set up a Linux VM or use WSL on Windows.

Version Control & Git

Git is the foundation of all DevOps workflows. Master: branching strategies (GitFlow, trunk-based), merge vs rebase, conflict resolution, Git hooks. Understand branching strategies that teams actually use. Learn to write meaningful commit messages. Understand how Git workflows connect to CI/CD — every push triggers a pipeline.

Containers: Docker

Docker is the cornerstone of modern DevOps. Learn: Dockerfiles (multi-stage builds), Docker Compose for multi-container apps, image optimization (Alpine base, layer caching), container networking, volume management. Practice: containerize a web app + database + Redis setup. Understand why containers exist — consistency between development and production environments. See our Docker beginner guide.

CI/CD Pipelines

GitHub Actions: Start here — it's free and integrated with your repos. Build pipelines that: lint code, run tests, build Docker images, push to registry, deploy to staging/production. GitLab CI: Strong alternative with built-in container registry. Jenkins: Legacy but still widely used. Key concepts: build stages, artifacts, environment variables, secrets management, deployment strategies (blue-green, canary, rolling). Learn GitHub Actions CI/CD first.

Container Orchestration: Kubernetes

Kubernetes manages containers at scale. Core concepts: Pods, Deployments, Services, Ingress, ConfigMaps, Secrets, Namespaces. Learn: kubectl commands, YAML manifests, Helm charts for packaging. Start with Minikube or kind for local development. Then move to managed services: EKS (AWS), GKE (Google), AKS (Azure). K8s has a steep learning curve — budget 2-3 months of focused learning.

Infrastructure as Code

Terraform: Declare infrastructure in HCL, apply changes idempotently. Supports all major clouds. Learn: providers, resources, modules, state management, workspaces. Pulumi: IaC using real programming languages (TypeScript, Python). See our Terraform vs Pulumi comparison. Ansible: Configuration management for existing servers. Start with Terraform for infrastructure provisioning, add Ansible for configuration if needed.

Monitoring & Observability

Metrics: Prometheus + Grafana for dashboards. Monitor CPU, memory, disk, request latency, error rates. Logging: ELK stack (Elasticsearch, Logstash, Kibana) or Loki + Grafana. Tracing: Jaeger or Zipkin for distributed request tracing. Alerting: PagerDuty, OpsGenie for incident notification. The three pillars — metrics, logs, traces — give complete visibility into production systems. Learn observability fundamentals.

Cloud Platforms

Pick one cloud to start: AWS (largest market share), GCP (best for K8s and AI), Azure (dominant in enterprise). Learn core services: compute (EC2/VMs), storage (S3), databases (RDS), networking (VPC), IAM. Get certified — AWS Solutions Architect Associate is the most recognized DevOps certification. The goal is understanding cloud primitives, not memorizing services.