[Presentation Report] I gave a talk on the theme "Introduction to RAG with Amazon Bedrock"

Introduction

Hello, I'm Kamino from the consulting department who loves the supermarket La Mu.

Recently, I presented at a customer seminar on the theme "Introduction to RAG with Amazon Bedrock"! (The official seminar title included a catchphrase, but in this article I've included only the core part).

This time, I explained simple content to introduce you to Bedrock and RAG!

Presentation Materials

Here are the presentation materials.

Comments

For those who are considering implementing generative AI and RAG using AWS but don't know where to start, I thought it would be important to first grasp the overall picture of Bedrock and RAG, which is why I explained this theme.

The content covers everything from an overview of Amazon Bedrock to the mechanics of RAG, and how to apply it to actual business operations, all targeted at beginners. I introduced use cases in two stages: simple in-house document search/FAQ responses, and RAG that investigates across multiple information sources.

As I mention in the materials, RAG is not something you just build and finish - it's important to cycle through verification and improvement. I believe that by organizing documents, creating expected Q&As, actually evaluating them, identifying bottlenecks, and tuning accordingly, we can gradually bring the quality of RAG answers closer to the target level through this repetition. (I'm writing this simply, but there are of course many difficult areas, so it's important to diligently work through these cycles...!)

Conclusion

I hope that after seeing these presentation materials, you feel motivated to try using Bedrock and creating RAG!

Thank you very much for reading until the end!!