AWS profession - Use Case 1: Enterprise Document Q&A (RAG)

Business Problem

A company has thousands of internal PDFs containing HR policies, legal documents, engineering manuals, and SOPs.

Employees ask:

"How many vacation days can I carry forward?"

The LLM should answer only from company documents.

Architecture

                User

                  │

           API Gateway

                  │

              Lambda

                  │

          Amazon Bedrock

                  │

        Knowledge Base

                  │

        Vector Database

              │

      S3 Document Store

Flow

PDFs uploaded to S3

Bedrock Knowledge Base automatically

parses documents

chunks text

generates embeddings

stores embeddings

User asks question

API Gateway invokes Lambda

Lambda calls Bedrock

Bedrock retrieves relevant chunks

Foundation model generates answer

Answer returned

AWS Services

Service Purpose

S3 Document storage

Bedrock Knowledge Base RAG

Bedrock Foundation Model Answer generation

Lambda Business logic

API Gateway REST endpoint

IAM Permissions

CloudWatch Logs

Why This Architecture?

Advantages

No model training

Hallucination reduced

Documents stay private

Serverless

Easy scaling

Common Exam Question

Why not fine-tune?

Answer:

Because company documents change frequently.

Knowledge Bases update automatically while fine-tuning requires retraining.