What You'll Learn

13 Chapters · 14 Lessons

    1. Free Preview Free preview

    1. (Included in full purchase)

      Working Backwards

    1. (Included in full purchase)

      Understanding Agentic AI and Autonomous Agents

    1. (Included in full purchase)

      AWS Agentic AI Ecosystem

    1. (Included in full purchase)

      Designing the First Agent System

    1. (Included in full purchase)

      Building and Deploying Agents on Bedrock

    1. (Included in full purchase)

      Multi-Agent Orchestration

    1. (Included in full purchase)

      Integrating Tools, APIs, and External Systems

    1. (Included in full purchase)

      Agent Architecture Security: Building Secure Agents

    1. (Included in full purchase)

      Observability, Evaluation, and Human-in-the-Loop

    1. (Included in full purchase)

      Scaling to Production: High Availability and Cost Optimization

    1. (Included in full purchase)

      Advanced Optimization and Organizational Transformation

    1. (Included in full purchase)

      Emerging Trends and Future Directions

    2. (Included in full purchase)

      Index

About the Course

Don't Build Another AI Demo. Build an Agent Enterprise Can Depend On. Enterprises are moving from AI that answers questions to AI that does the work. Ultimate AWS Agents for Production AI Systems shows how to build, deploy, and operate autonomous agents that reason, plan, call tools, and act using AWS's purpose-built stack — Amazon Bedrock for inference, Amazon Bedrock AgentCore for production infrastructure, and the Strands Agents SDK for building them. The hard part is not the first prototype. It is everything after. The book is organized as the journey a production agent actually takes. You start by identifying an opportunity worth building with a Working Backwards PR/FAQ, then learn the agent loop and the AWS services that implement it. You design and deploy a single agent with Strands on AgentCore Runtime, then scale into multi-agent orchestration with router, graph, and hybrid patterns. From there the book turns to production — integrating tools through MCP and Knowledge Bases, enforcing tool-level access control and domain boundaries, instrumenting observability with evaluation and human review, and hardening for high availability, tiered models, prompt caching, and canary rollouts. By the end you will design, secure, operate, and cost-manage agent systems that carry real enterprise workloads, and lead the organizational shift that keeps them running.

About Your Instructor

Dr. Alicja Kwasniewska leads AI for Advertising and Marketing North America at AWS, specializing in a wide range of AI, ML, DL, and GenAI solutions across various data modalities. With 40+ peer-reviewed publications in AI (h-index: 14), she advises enterprise customers on real-time bidding, brand recognition, and AI-powered content generation. Brian Maguire is a Principal Solutions Architect at Amazon Web Services, helping customers bring AI and agentic systems to production. He was the VP of Engineering at Vantage Labs, and led teams building distributed ML and analytics platforms. Brian is a Wharton CTO Program graduate and co-author of Scalable Data Streaming with Amazon Kinesis. Avichal Chum is a Senior Solutions Architect at Amazon Web Services specializing in Artificial Intelligence (AI) and cloud computing. He designs scalable AWS solutions spanning compute, storage, Machine Learning (ML), and serverless architectures, and publishes on cloud best practices as well as autonomous AI systems for demanding production workloads. Sudeesh Sasidharan is a builder and Solutions Architect at Amazon Web Services who specializes in AI, agentic systems and cloud-native solutions for enterprise customers.​​ Sudeesh also serves on America On Tech’s AI Council, where he supports initiatives to expand access to AI and technology careers for students and emerging professionals from diverse backgrounds.