RAG

Serverless RAG on AWS

A document Q&A system on AWS, deployed end to end as infrastructure as code.

The problem

Companies want a document assistant inside their own AWS account, without servers to babysit.

How it works

Files dropped in S3 trigger an ingestion Lambda that extracts, chunks and embeds them with Titan Embeddings v2 into PostgreSQL with pgvector. A query Lambda behind API Gateway retrieves context and answers with Claude on Amazon Bedrock. The whole stack is defined in AWS CDK.

Pipeline

01S3 upload
02Ingestion Lambda
03Titan embeddings
04pgvector
05Claude on Bedrock

Highlights

Entire infrastructure defined in AWS CDK
Secrets Manager and least-privilege IAM roles
PDF, DOCX and TXT ingestion
Designed to run within the AWS Free Tier

Stack

AWS CDKLambdaBedrock · ClaudeTitan EmbeddingspgvectorAPI Gateway

Want something like this for your company?

I build production versions of these systems on your data and your infrastructure.

Related service: Knowledge assistants

Book a call