How jarodsim/event-driven-platform works
Project type and purpose
• This is a serverless, event-driven, distributed order-processing platform on AWS, orchestrated and versioned with Terraform.
• Its purpose is to receive HTTP requests, publish them as events on an EventBridge bus, and fan-out those events to multiple independent Lambda consumers (validation, invoicing, notification), each with its own SQS dead-letter queue.
File structure
• Top-level “infra/” folder: all Terraform code.
• infra/modules: reusable modules for AWS API Gateway, Lambda, EventBridge, SQS, IAM roles/policies.
• infra/services: four service definitions (create_order, validate_order, invoice_order, notificate_order), each provisioning its own Lambda, EventBridge rule, and DLQ.
• infra/envs/dev: environment-specific tfvars.
• .github/workflows: CI/CD pipelines for Terraform.
• “lambdas/”: service code directory with one subfolder per Lambda.
• readme.md: describes event-driven architecture, flowchart, components, and AWS stack.
Key takeaways from the README
• Core flow: API Gateway → Entry Lambda → EventBridge bus → three processing Lambdas → SQS DLQs for failures
• Tech: AWS API Gateway, Lambda, EventBridge, SQS DLQs, CloudWatch, IAM, Terraform, Python