AI Workshop (Chatbot)

This workshop provides a detailed guide on how to build a Product Consultation Chatbot using RAG (Retrieval-Augmented Generation) architecture on the AWS platform.

1. System Architecture

The system utilizes the following AWS services:

• Amazon Bedrock: Provides AI models (LLMs).
  • Generation Model: Amazon Nova Lite (anthropic.claude-3-haiku-20240307-v1:0) for answering questions.
  • Embedding Model: Cohere Embed Multilingual (cohere.embed-multilingual-v3) for data vectorization.
• Amazon RDS (PostgreSQL): Stores product data and vectors (using the Dbeaver extension).
• AWS Lambda: Intermediate logic processing function (Serverless backend).