A self-paced, project-based learning path for anyone who wants to build real AWS and DevOps skills from scratch. Each project is a standalone, deployable exercise with step-by-step instructions, working code, and cleanup scripts.
Who this is for: Software engineers, sysadmins, or students who learn best by doing — not by watching videos. No prior AWS experience needed to start; by the end you will have hands-on practice with every major AWS service used in professional engineering roles.
What you will have built after all 20 projects:
- Deployed, globally distributed websites and APIs
- Automated infrastructure provisioned with Terraform
- Containerised applications running on ECS and EKS
- CI/CD pipelines with GitHub Actions
- Centralised logging, metrics, and alerting stacks
- Security automation, chaos experiments, and cost optimisation workflows
See QUICK_START.md for environment setup (AWS CLI, billing alerts, prerequisites).
New to AWS? Start at Project 1 and work forward in order.
Some AWS experience? Jump to the level that matches your background using the table below.
Experienced? Projects 12–20 cover Kubernetes, GitOps, observability, chaos engineering, and FinOps.
Foundation projects covering core AWS services.
| # | Project | Key Services | Time |
|---|---|---|---|
| 1 | Static Website on S3 + CloudFront | S3, CloudFront, ACM, Route 53 | 1–2 hrs |
| 2 | Linux Server Setup on EC2 | EC2, VPC, Security Groups, Nginx | 2–3 hrs |
| 3 | Serverless Contact Form | Lambda, API Gateway, SES | 2–3 hrs |
| 4 | RDS Database with Backups | RDS, Multi-AZ, CloudWatch | 2–3 hrs |
| 5 | CI/CD Pipeline | GitHub Actions, OIDC, S3 | 2–3 hrs |
Infrastructure as Code, containers, and observability fundamentals.
| # | Project | Key Services | Time |
|---|---|---|---|
| 6 | Infrastructure as Code | Terraform, VPC, EC2, RDS modules | 3–4 hrs |
| 7 | Containerise App | Docker, ECS, ECR, ALB | 3–4 hrs |
| 8 | Centralised Logging | CloudWatch, CloudWatch Agent | 3–4 hrs |
| 9 | Secrets Management | Secrets Manager, Lambda rotation | 3–4 hrs |
| 10 | Auto Scaling Web Tier | ALB, ASG, Launch Templates | 3–4 hrs |
| 11 | Event-Driven Data Pipeline | SQS, Lambda, S3, DynamoDB | 3–4 hrs |
Kubernetes, GitOps, observability, and enterprise deployment patterns.
| # | Project | Key Services | Time |
|---|---|---|---|
| 12 | Kubernetes on EKS | EKS, Helm, HPA, PDB | 4–6 hrs |
| 13 | GitOps with ArgoCD | ArgoCD, Helm, App-of-Apps | 4–6 hrs |
| 14 | Full Observability Stack | Prometheus, Grafana, Loki | 4–6 hrs |
| 15 | Blue/Green & Canary Deployments | CodeDeploy, ALB, ECS | 4–6 hrs |
| 16 | AWS Security Posture | GuardDuty, Security Hub, Config | 4–6 hrs |
Multi-region architecture, platform engineering, chaos testing, and cost governance.
| # | Project | Key Services | Time |
|---|---|---|---|
| 17 | Multi-Region Active-Active | Global Accelerator, DynamoDB Global Tables | 6–8 hrs |
| 18 | Platform Engineering | Backstage, Terraform modules, IDP | 6–8 hrs |
| 19 | Chaos Engineering | AWS FIS, Chaos Mesh, steady-state monitoring | 6–8 hrs |
| 20 | FinOps Cost Optimisation | Cost Explorer, Budgets, tagging SCPs | 6–8 hrs |
AWS Project N - Name/
├── README.md # Step-by-step guide, architecture, cleanup, troubleshooting
├── deploy*.sh # Automated deployment (Projects 1–7)
├── cleanup*.sh # Resource teardown
└── ... # Application code, configs, Helm charts, Terraform, etc.
Projects 1–7 include fully automated deploy/cleanup scripts — run one command and the infrastructure is live. Projects 8–20 are README-driven: all CLI commands are complete and copy-pasteable.
# Automated projects (1–7)
cd "AWS Project 2 - Linux Server Setup on EC2"
bash deploy.sh
# ... test, learn, then:
bash cleanup.sh
# README-based projects (8–20)
cd "AWS Project 8 - Centralized Logging"
# Open README.md and follow the stepsSee DEPLOYMENT_GUIDE.md for full deployment methods and APPLICATIONS_SUMMARY.md for a complete list of every file in every project.
- Always run cleanup scripts after finishing a project — leaving resources running is the most common source of unexpected charges
- Most projects cost under $5 if cleaned up within 24 hours
- Set a billing alarm before starting (see QUICK_START.md)
- Default region is
us-east-1— change withaws configure set region YOUR_REGION
| Domain | Topics |
|---|---|
| Core AWS | EC2, S3, RDS, Lambda, VPC, IAM, CloudFront |
| Infrastructure as Code | Terraform modules, state management |
| Containers | Docker, ECS, ECR, EKS, Kubernetes |
| CI/CD | GitHub Actions, CodeDeploy, GitOps, ArgoCD |
| Observability | CloudWatch, Prometheus, Grafana, Loki |
| Security | GuardDuty, Security Hub, Secrets Manager, Config |
| SRE | Auto scaling, chaos engineering, SLI/SLO |
| FinOps | Cost Explorer, budgets, tagging, right-sizing |
Completing these projects builds hands-on experience directly relevant to:
- AWS Certified Solutions Architect — Associate
- AWS Certified Developer — Associate
- AWS Certified SysOps Administrator — Associate
- AWS Certified DevOps Engineer — Professional
Start with Project 1 or jump to whichever level matches your experience.