Build an ecommerce product recommendation chatbot with Amazon Bedrock Agents

This article explains how to build an ecommerce product recommendation chatbot using Amazon Bedrock Agents and foundation models from Amazon Bedrock.

Specifically, the article covers:

• Solution overview: Creating an intelligent chatbot that can converse naturally with users, understand their gift preferences, query a product catalog API, and recommend relevant products
• Prerequisites: Required AWS services, IAM policies, and permissions
• Deploying solution resources: Using AWS CloudFormation to create a DynamoDB table for product data, a Lambda function for the product API, and a Lambda function to populate sample product data
• Creating the agent: Requesting model access, creating the agent and action group using CloudFormation, understanding the agent instructions and OpenAPI schema
• Testing the chatbot: Example conversations demonstrating how the chatbot gathers preferences and recommends products, along with rationale explanations
• Clean up: Deleting the CloudFormation stacks to remove resources