Build Production-Grade Multi-Agent AI

About this course

Are you ready to move beyond basic chatbots and build enterprise-ready, autonomous AI systems? This course provides a comprehensive blueprint for developing production-grade multi-agent Generative AI applications on AWS. Using a real-world "AnyCompany Coffee Shop" case study, you will learn how to design an AI Barista that doesn't just chat, but actually executes business logic.

We begin with Strategy and Ideation, where you will use Streamlit to model the financial impact and ROI of your AI agents, ensuring your projects deliver clear business value. From there, we dive into a robust Data Strategy, teaching you how to leverage your existing data—including Amazon DynamoDB for orders and Amazon Aurora for store operations—as a competitive differentiator.

The heart of the course focuses on the technical implementation of Multi-Agent Collaboration (MAC) architecture. You will learn to use the Strands SDK to build a Supervisor Agent that orchestrates specialized sub-agents for Orders, Menu, Payments, Stores, and Promos. You will also master Retrieval-Augmented Generation (RAG) by building Knowledge Bases using Amazon Bedrock and Amazon OpenSearch Serverless to ensure your agents provide accurate, policy-compliant responses.

By the end of this course, you will have a deep understanding of the Agentic Loop, Harness and Loop Engineering—how models use tools and prompts to autonomously plan, reason, and take actions to solve complex tasks. This is the definitive guide for developers looking to master the next frontier of AWS cloud architecture and Generative AI.

Prices, discounts and availability are set by Udemy. We may earn a commission when you purchase through links on this site.