This project uses Amazon Bedrock, AWS Lambda, API Gateway, and Amazon S3 to generate production-ready Dockerfiles using Gen-AI models. Everything is deployed using Terraform Cloud.

• Generate best-practice Dockerfiles for any programming language using Amazon Bedrock (Llama3).
• Store the generated Dockerfile in S3.
• Return a presigned URL to download the file.
• Full infrastructure-as-code using Terraform Cloud.

Here's how the whole system works:

1. Postman sends a POST request to API Gateway with the programming language.
2. API Gateway triggers the Lambda Function.
3. Lambda Function:
   • Sends a prompt to Amazon Bedrock
   • Receives Dockerfile content
   • Stores the Dockerfile in Amazon S3
   • Generates a presigned S3 URL
   • Returns the URL back to API Gateway
4. API Gateway responds to Postman with the final download URL.

The Lambda function:

• Accepts { "language": "python" } as input
• Generates a Dockerfile using Bedrock
• Saves it to S3
• Returns a presigned download URL

See lambda/app.py for details.

• Amazon Bedrock - Gen-AI model invocation
• AWS Lambda - Serverless backend
• Amazon S3 - File storage
• API Gateway - HTTP interface
• Terraform Cloud - Infra provisioning

Read the full blog post with prompt debugging lessons, Bedrock quirks, and architecture breakdown here:
👉 Medium Blog

MIT

Zlash65
GitHub