Learn AWS DevOps the right way: 100% hands-on, 100% practical, 100% free (almost).
This is not another AWS course with boring slides and theory. This is 7 days of intense, hands-on learning where you build real infrastructure on AWS using production-grade tools.
By Day 7, you'll have:
- Deployed a complete microservices application on AWS
- Mastered Terraform, Ansible, Jenkins, Docker, and Kubernetes
- Built a full observability stack (Prometheus, Grafana, OpenTelemetry)
- Created a portfolio of 7 real-world projects
- Understood AWS architecture at Solutions Architect level
And the best part? Total cost: ~$20-25 for the entire 7 days (if you keep everything running 24/7, less if you clean up daily).
I'm Koti, a DevOps Engineer at TransUnion managing 1000+ Kubernetes workloads across 50+ clusters. I've made every mistake possible in AWS so you don't have to.
What makes this different:
- ✅ No application development - We use existing open-source apps (Sock Shop microservices)
- ✅ No toy projects - Everything is production-grade and follows AWS Well-Architected Framework
- ✅ No theoretical BS - 100% hands-on, real AWS Console and CLI work
- ✅ No confusion - Crystal clear instructions, tested personally by me
- ✅ No hidden costs - Complete transparency on every dollar spent
You're ready for this if you:
- Have basic Linux knowledge (our Linux course level)
- Know basic shell scripting (our Shell Scripting course level)
- Understand what Docker containers are (don't need to be an expert)
- Have used Git before
- Want to learn AWS the practical way, not the theoretical way
You'll struggle if you:
- Have never touched a terminal
- Expect copy-paste without understanding
- Want quick certifications without depth
- Aren't willing to spend 7-8 hours per day
Theme: Understanding AWS the Hard Way
What you'll build:
- Custom VPC with public and private subnets across 2 AZs
- EC2 instance running Docker with Sock Shop microservices (14 containers)
- RDS MySQL database in private subnet
- Application Load Balancer for traffic distribution
- CloudWatch monitoring, alarms, and billing alerts
What you'll learn:
- AWS networking fundamentals (VPC, subnets, routing, security groups)
- EC2 instance management
- RDS managed databases
- Load balancing and high availability concepts
- Cost monitoring and Free Tier management
Time: 7-8 hours Cost: ~$0.50 (ALB only)
Theme: Automate Everything You Did Yesterday
What you'll build:
- Terraform modules for VPC, EC2, RDS, ALB, Security Groups
- Reusable infrastructure code
- Remote state management in S3 with DynamoDB locking
- Multi-environment setup (dev/staging/prod) with workspaces
What you'll learn:
- Terraform fundamentals and best practices
- Infrastructure as Code philosophy
- State management and why it matters
- Creating reusable modules
- Destroying and recreating infrastructure in minutes
Time: 7-8 hours Cost: ~$0.50/day (same infrastructure, now automated)
Theme: Configure at Scale
What you'll build:
- Ansible playbooks for automated Docker installation
- Application deployment automation
- Database configuration as code
- Dynamic inventory from AWS
- Ansible Vault for secrets management
What you'll learn:
- Ansible fundamentals
- Configuration vs Infrastructure (Terraform vs Ansible)
- Idempotent operations
- Role-based playbook structure
- Secrets management
Time: 7-8 hours Cost: ~$0.50/day
Theme: Automate the Automation
What you'll build:
- Jenkins server on EC2
- Multi-stage CI/CD pipeline (Terraform → Ansible → Deploy → Test)
- GitHub integration
- Blue-green deployment strategy
- Automated rollback mechanisms
What you'll learn:
- CI/CD principles
- Pipeline as Code (Jenkinsfile)
- Deployment strategies
- GitOps workflow
- Integration with AWS services
Time: 7-8 hours Cost: ~$1.00/day (additional EC2 for Jenkins)
Theme: Kubernetes in the Cloud
What you'll build:
- EKS cluster (or ECS Fargate for free tier)
- Kubernetes manifests for Sock Shop
- Horizontal Pod Autoscaling
- Cluster Autoscaling
- Ingress configuration with ALB
What you'll learn:
- Kubernetes architecture and concepts
- EKS vs ECS vs self-managed K8s
- Container orchestration patterns
- Auto-scaling configurations
- Cloud-native application deployment
Time: 7-8 hours Cost: ~$3.50/day (EKS control plane) OR $0 with ECS Fargate
Theme: See Everything, Know Everything
What you'll build:
- Prometheus for metrics collection
- Grafana with custom dashboards
- AlertManager for intelligent alerting
- OpenTelemetry Collector for distributed tracing
- Complete observability across all services
What you'll learn:
- The three pillars: Metrics, Logs, Traces
- Prometheus query language (PromQL)
- Building effective dashboards
- Alert fatigue prevention
- Production debugging techniques
Time: 7-8 hours Cost: ~$3.50/day (same as Day 5)
Theme: Making It Real
What you'll build:
- Multi-AZ high availability setup
- Auto Scaling Groups
- Disaster recovery plan and runbook
- Security hardening (IAM policies, Secrets Manager)
- Cost optimization analysis
What you'll learn:
- Production vs Development mindset
- AWS Well-Architected Framework
- Security best practices
- Disaster recovery planning
- Cost optimization strategies
Time: 7-8 hours Cost: ~$3.50/day
| Day | Services | Daily Cost | Notes |
|---|---|---|---|
| 1 | EC2 + RDS + ALB | ~$0.50 | Free Tier for EC2/RDS |
| 2 | Same + S3/DynamoDB | ~$0.50 | S3/DynamoDB free |
| 3 | Same | ~$0.50 | No new services |
| 4 | Add Jenkins EC2 | ~$1.00 | Extra t2.micro |
| 5 | Add EKS/ECS | ~$3.50 | EKS control plane OR $0 with ECS |
| 6 | Add monitoring | ~$3.50 | Runs on K8s |
| 7 | Same | ~$3.50 | Optimization day |
Total for 7 days (running 24/7): ~$20-25
- Clean up daily: Run cleanup scripts at end of each day
- Use ECS instead of EKS on Day 5 (saves $15/week)
- Stop instances when not using: EC2 and RDS can be stopped
- Delete and recreate: With Terraform (Day 2+), recreation takes 5 minutes
Realistic cost if you clean up daily: ~$10-15 for the entire week
- ✅ Basic Linux commands (cd, ls, vim, grep, find)
- ✅ Basic shell scripting (variables, loops, functions)
- ✅ SSH and working with remote servers
- ✅ Git basics (clone, commit, push)
- ✅ What Docker containers are (conceptually)
Test yourself:
# If you can do these, you're ready
ls -la
grep "error" /var/log/syslog
ssh user@server
git clone https://github.com/example/repo
docker psAWS Account:
- AWS Free Tier account (we'll set up on Day 1)
- Credit/debit card (AWS requires this, won't charge if we stay in limits)
Your Computer:
- Terminal (Mac/Linux) or WSL (Windows)
- 8GB RAM minimum
- 20GB free disk space
- Stable internet connection
Tools to Install:
- AWS CLI
- Git
- Text editor (VS Code recommended)
- SSH client
We'll install Terraform, Ansible, Docker, kubectl as we need them.
aws-devops-7days/
├── README.md # This file
├── day1-foundations/ # Day 1: Manual deployment
│ ├── README.md # Complete Day 1 guide (1800+ lines)
│ ├── QUICK-REFERENCE.md # Quick commands reference
│ ├── manual-steps/ # Step-by-step guides (7 detailed docs)
│ │ ├── 01-aws-account-setup.md
│ │ ├── 02-vpc-setup.md
│ │ ├── 03-ec2-setup.md
│ │ ├── 04-sockshop-deployment.md
│ │ ├── 05-rds-setup.md
│ │ ├── 06-alb-setup.md
│ │ └── 07-monitoring-setup.md
│ ├── scripts/ # Helper & automation scripts
│ │ ├── setup-day1.sh # Automated infrastructure setup
│ │ ├── verify-deployment.sh
│ │ ├── check-costs.sh
│ │ └── cleanup-day1.sh
│ └── diagrams/ # Architecture diagrams
├── day2-terraform/ # Day 2: Terraform IaC
├── day3-ansible/ # Day 3: Ansible configuration
├── day4-jenkins/ # Day 4: CI/CD pipeline
├── day5-kubernetes/ # Day 5: EKS/ECS deployment
├── day6-observability/ # Day 6: Monitoring stack
├── day7-production/ # Day 7: Production hardening
└── common/ # Shared resources
├── cleanup-scripts/
└── troubleshooting.md
- Read the day's README completely before starting
- Follow along step-by-step - don't skip ahead
- Type every command yourself - no copy-paste without understanding
- Take notes on what you learn
- Complete the day's project fully before moving on
- Run cleanup scripts if you want to save money
- Document issues you face for future reference
- Each day: 7-8 hours (including breaks)
- Total course: 50-55 hours of content
- Can spread: 2-3 days per section if needed
My recommendation: Dedicate focused time blocks. Don't rush. Quality over speed.
- ✅ Manual AWS deployment of microservices app
- ✅ Complete Terraform infrastructure codebase
- ✅ Ansible configuration management automation
- ✅ Full CI/CD pipeline with Jenkins
- ✅ Kubernetes/ECS production deployment
- ✅ Observability stack with custom dashboards
- ✅ Production-ready, secure, cost-optimized infrastructure
- ✅ AWS services (VPC, EC2, RDS, ALB, EKS/ECS, S3, IAM, CloudWatch)
- ✅ Infrastructure as Code (Terraform)
- ✅ Configuration Management (Ansible)
- ✅ CI/CD (Jenkins)
- ✅ Containers (Docker)
- ✅ Orchestration (Kubernetes/ECS)
- ✅ Monitoring (Prometheus, Grafana, OpenTelemetry)
- ✅ Security (IAM, Security Groups, Secrets Management)
- ✅ Cost Optimization
You'll be able to answer questions like:
- "Explain your AWS networking setup"
- "How do you deploy applications to production?"
- "Walk me through your CI/CD pipeline"
- "How do you monitor and debug production issues?"
- "How do you manage infrastructure as code?"
If you're stuck:
- Check the troubleshooting section in each day's README
- Review the Quick Reference guides
- Search GitHub Issues (someone probably faced the same issue)
- Open a new issue with:
- What you're trying to do
- What command you ran
- The exact error message
- Screenshots if applicable
Found a typo? Have a better way to explain something? Want to add content?
- Fork the repository
- Make your changes
- Submit a Pull Request
- I review and merge
All contributions are welcome!
I'd love to see your progress!
- Tag me on LinkedIn: [Your LinkedIn]
- Use hashtag: #AWSDevOps7Days
- Share your dashboards, architectures, learnings
I'm Koti, a DevOps Engineer at TransUnion where I manage:
- 1000+ production Kubernetes workloads
- 50+ Kubernetes clusters
- Multi-cloud infrastructure (AWS, Azure, GCP)
- Platform engineering for 200+ developers
Background:
- 6+ years in DevOps and Cloud Engineering
- PyCon India 2025 Core Organizer
- Speaker at various tech conferences
- Passionate about sharing practical knowledge
Why I created this: I was frustrated with AWS courses that:
- Focus on certifications, not skills
- Use toy projects, not real infrastructure
- Cost hundreds of dollars
- Leave you unprepared for real work
So I built what I wish existed when I was learning.
This section will be updated as people complete the course. Be the first!
No. This course teaches you practical AWS skills. If you want certifications later, this will make studying much easier.
Yes! Each day is self-contained. You can do 1 day per weekend and finish in 7 weeks.
Set up billing alerts on Day 1 (we walk through this). Monitor costs daily. Clean up resources when not using them. Realistic total: $10-25.
No, this course is AWS-specific. The concepts transfer, but commands and services don't.
Yes! This covers:
- 70% of AWS Solutions Architect Associate
- 60% of AWS DevOps Engineer Professional
- 50% of AWS SysOps Administrator
But this focuses on practical skills, not exam tricks.
Absolutely! You'll have:
- Public GitHub repository with all code
- Working knowledge of production AWS infrastructure
- Real projects you can demo in interviews
MIT License - Use this freely. Learn. Share. Give credit where it's due.
- WeaveWorks for the Sock Shop microservices demo application
- HashiCorp for Terraform
- Red Hat for Ansible
- AWS for Free Tier and great documentation
- Every DevOps engineer who shared their knowledge openly
Stop reading about DevOps. Start doing DevOps.
👉 Begin Day 1: AWS Foundations
See you on the other side, where you're an AWS DevOps engineer who actually knows what they're doing.
-Koti
Star this repo if you find it useful! Share it with someone learning DevOps! Follow the journey - more content coming soon!