
Build a production AWS infrastructure for 8 microservices using Terraform, EKS, Argo CD, Helm, and GitHub Actions, with service discovery, API gateway, and observability powered by agentic AI.
Verify prerequisites for this course: DevOps fundamentals including Docker, Kubernetes, Terraform, AWS Cloud, and Linux; familiarize agentic AI concepts like skills, sub-agent, and MCP server.
Explore your companion repository guiding this course, with sections containing article.md and prompt.md. Read the readme and resource.md to learn tools and Jira stories, then use prompts to generate resources.
Become a devops engineer and go live on AWS by building infrastructure with Terraform, deploying eight Spring Boot services to EKS, and enabling GitOps with Argo CD.
Walk through the eight-service microservices architecture, highlighting config server, Eureka discovery, API gateway, MySQL-backed services, GenAI with vector store, and the DevOps stack (Helm, Argo CD, Terraform).
Install cloud code, a context-aware cli agent that runs in your terminal to scan your codebase, write real Terraform modules, and run commands with project context.
cloud code reads the app repo for services, ports, mysql needs, and health endpoints, then consults the platform repo and cloud.md to map infrastructure and build order.
Learn how to manage aws costs and stay under $10 by using the free tier options, destroying infrastructure after each session, and understanding eks control plane fees.
Run all eight microservices locally with Docker Compose to see the pet clinic in action. Set up cloud code and begin Terraform with remote state using S3 and DynamoDB.
Run eight microservices with docker compose up -d, enforcing startup order via config and discovery servers and health checks, with api gateway and observability through Eureka, Zipkin, Prometheus, and Grafana.
Master the Jira board workflow to pick epics and tickets, read the technical spec, and drive cloud code builds from Terraform foundations to remote state and GitOps with Argo CD.
Connect Claude Code to Jira via Atlassian mcp, authenticate with OAuth, and generate five pet clinic platform stories (directory structure, bootstrap, backends, providers, versions) with terraform init, validate, and fmt.
Explore building a full public VPC with two public subnets across two availability zones, an internet gateway, and security groups, mapping public IPs for EKS, dev, and prod.
Compare the all-public-subnet production design with the learning setup, using security groups as the perimeter, ADR-0001, and cost savings from no nat gateway.
Build a complete reusable vpc module with four security groups, subnets, and igw in dev and prod using cloud code. Experience agentic workflow with automatic review, terraform validate, and hooks.
Execute a safe terraform deployment workflow by planning with plan dev, reviewing the plan, and applying to create real aws resources, handle errors, and store state in s3.
Build and configure an Amazon EKS cluster with managed node groups and OIDC integration using IRSA, including Core DNS, kube proxy, VPC CNI, and the EBS CSI driver.
Build the entire EKS module with Cloud Code, implementing cluster and IAM roles, node groups, kubectl access, and add-ons, then validate and review for security and best practices.
Review and deploy the EKS module by validating IAM trust policies, OIDC/IRSA setup, and add-on versions, then run terraform plan and apply to provision resources.
Leverage a Terraform module to build eight private ECR repositories for dev and prod, enabling scan on push, lifecycle policies, and tag immutability, with ARM64 docker images for Graviton EKS.
Verify eight ECR repositories in the petclinic-dev-namespace, confirm v1.0.0 images and scan-on-push results, and assess CVEs to maintain a secure, repeatable build workflow.
Set up a shared RDS MySQL on db.t4g.micro for the pet clinic apps, with Terraform-generated credentials in Secrets Manager and secure access from the EKS cluster.
Review the RDS module and credential flow, ensure security group, parameter group, and lifecycle settings align across dev and prod, then deploy and collect the RDS endpoint.
Verify connectivity from a debug pod inside the cluster to the RDS MySQL instance, validate credentials from Secrets Manager, and confirm utf-8mb4 parameter group is applied.
Build the entire DNS and ingress stack for EKS using Terraform and Helm. Configure Route 53 hosted zone, ALB ingress, DNS records, and dev environment wiring with validation.
Review and deploy the dns, certificate validation, and ingress settings across two phases, ensuring proper data source hosted zone usage, acm validation, and accurate ingress annotations for a secure alb.
Verify the infrastructure by performing four checks: DNS resolution via Route 53, ACM certificate validation, ALB ingress, and HTTPS access in the browser; implemented with Terraform, Helm, and Kubernetes.
Explore end-to-end secrets management with AWS Secrets Manager, IRSA, and the External Secrets Operator, as secrets flow from AWS to Kubernetes secrets and into pod environments.
Verify end-to-end secret management from AWS secrets manager to Kubernetes, ensuring ESO health, synced external secrets, and decoded values match across RDS credentials and OpenAI API key.
Deploy eight services to Kubernetes using base manifests and overlays, with four files per service, init containers, and health probes ensuring startup order from config server to discovery server.
Verify that all eight pods and services are up, reachable, and end-to-end functional from browser to database through the ALB, API gateway, Eureka, and RDS.
Explain helm chart anatomy and why helm, replacing yaml files with templates and helpers. Describe the values merge order, chart defaults, per environment, then per service—and Argo CD's GitOps workflow.
Build with Claude Code by converting the plane yaml into a helm chart, templating deployments, services, and config maps, with per service and environment values, and validate with helm lint.
Review and deploy with helm upgrade --install, verify templates and values merge, then deploy in startup order from config server to all services, validating rds endpoints and secrets.
What if you could deploy production-grade AWS infrastructure without writing a single line of code yourself?
That’s exactly what this course is about. You’ll take a real Spring Boot microservices application—eight services, real databases, real traffic—and push it all the way to production on AWS. Every Terraform module, every Kubernetes manifest, every CI/CD pipeline, and every runbook is generated by Claude Code. Your role is to think like an architect: write precise prompts, review the outputs, and make sure everything is production-ready.
This isn’t a step-by-step tutorial. It’s a project.
You step into the role of a DevOps engineer handed a Jira board and expected to deliver. You’ll work through real epics—networking, compute, container registry, databases, secrets, GitOps, observability—in the same sequence a real production team would follow.
What you’ll build:
A VPC with public subnets across multiple availability zones
An Amazon EKS cluster running cost-optimized Graviton ARM nodes
Amazon RDS MySQL for persistent storage
Amazon ECR with lifecycle policies and vulnerability scanning
A GitOps pipeline using ArgoCD (auto-sync for dev, manual approvals for production)
GitHub Actions CI pipelines that build, push, and trigger deployments
Secrets Manager integrated with External Secrets Operator for Kubernetes
A full observability stack with Prometheus, Grafana, Fluent Bit, and Zipkin
Why Claude Code?
AI doesn’t replace engineers—it amplifies them. But only if you know how to guide it, evaluate its output, and catch what it misses. This course focuses on building that skill in the context of a real-world project, so you walk away with both working infrastructure and a repeatable workflow.
By the end, you’ll have:
A production-ready AWS platform in your GitHub portfolio
Hands-on experience with Terraform, EKS, ArgoCD, and GitHub Actions
A repeatable, AI-assisted workflow you can apply to future projects
If you’ve been meaning to get serious about cloud infrastructure, this is where it starts.