AI Agents on AWS: Beginner’s guide to building agents on AWS

Bunny Kaushik, Mona M

Build, scale, and deploy autonomous AI agents on AWS using Bedrock, memory, and multi-agent architectures for real-world systems

Key Features
• Progress from agentic AI concepts to production-ready deployment on AWS
• Design and build AI agents with tools, memory, and orchestration
• Deploy, evaluate, and govern agents using Amazon Bedrock AgentCore

Build reliable AI agents on AWS and move beyond prototypes with practical guidance. This book helps you design, deploy, and manage agentic systems that reason, use tools, and collaborate to solve real-world tasks.
–
You start with the foundations of agentic AI, understanding how agents think, act, and interact. Design single-agent systems, extend them with tool and function calling, and add memory for context-aware behavior using Amazon Bedrock and SageMaker AI. You will also use LangGraph and Strands to structure agent workflows.
–
As you progress, create multi-agent systems and orchestrate workflows where specialized agents collaborate. Learn how MCP and A2A enable communication, interoperability, and modular design across agent ecosystems.
–
Focus on production readiness by deploying and scaling agents with Bedrock AgentCore, evaluating performance, and implementing observability, monitoring, and governance. By the end, you will be able to design and operate robust AI agent systems on AWS.

Who this book is for
This book is for AI engineers, machine learning engineers, cloud architects, software engineers, and developers who want to build and deploy autonomous AI systems on AWS. It is also well suited for graduate and undergraduate students who want to enter the agentic AI space. Familiarity with Python, APIs, and basic AI/ML concepts is recommended, but the book is designed to make agentic AI accessible to readers who are still building their foundation.