From Idea to EKS in 2 hours, with Kiro
11:00 – 13:00
From idea to EKS in 2 hours, with Kiro
Most AI coding demos build a todo app. This one builds production-shaped infrastructure on real AWS.
In three hands-on hours you'll go from an empty folder to a Python API running on Amazon EKS, with its own ECR repository, its own DynamoDB table, EKS Pod Identity for secure AWS access, and a public Application Load Balancer. You'll write the application code, the Terraform that provisions your AWS resources, the Dockerfile, and the Kubernetes manifests, all using Kiro, AWS's spec-driven AI IDE. You'll also see Kiro debug a broken deployment by running kubectl and aws commands on its own, the way a colleague would.
By the end you'll have a working app on a public URL that you provisioned and shipped yourself, plus a working mental model for spec-driven platform engineering with AI tooling.
Audience: developers, DevOps and platform engineers, anyone curious about how AI tools fit into real Kubernetes workflows. No EKS experience needed; basic command-line comfort assumed.
Prerequisites: laptop with Docker, kubectl, AWS CLI v2, Terraform 1.5+, and Kiro installed.