aws-cdk-starter
aws-cdk-starter is a collection of reusable CDK stacks built by Vladimir Budilov to automate AWS resource provisioning across multiple services. Designed to standardize deployments that previously used a mix of Serverless, SAM, and Terraform, it focuses on building a data ingestion application using Kinesis, EKS, DynamoDB, and other AWS services. Current stacks include Cognito with Lambda triggers, DynamoDB tables for users and tickers, an Elasticsearch service with associated IAM roles, a Kinesis stream, an ECS Fargate service with IAM roles, ECR repositories, and Parameter Store integration for endpoint discovery. An S3 bucket for SPA hosting is planned. Deployment requires Java/Maven for building and the AWS CDK CLI for synthesis and deployment. Users first run mvn package and cdk synth after code changes, then cdk bootstrap once per account, followed by deploying the ECR stack to create a repository for container images, pushing a Docker image, and finally deploying the remaining stacks including DynamoDB