book

AI Agents on AWS

Name: AI Agents on AWS
ISBN: 9781806387212

by Bunny Kaushik, Mona M

April 2026

Intermediate

290 pages

7h 14m

English

Packt Publishing

Read now

Unlock full access

Preface
Free benefits with your book
Chapter 1: Understanding AI Agents on AWS
What is RAG and how does it work?What are agents?Agent capabilitiesExploring the AWS Agentic StackBuilding AI agents on AWSProgressing from LLMs to multi-agent systemsUnderstanding agent protocolsModel Context ProtocolAgent-to-Agent ProtocolIntroducing AWS Strands SDKBuilding a "Hello World" agentSetting up your environment in Amazon SageMaker studioCreating a simple agentSummary
Chapter 2: Building Agents with Tools
Technical requirementsWhy do agents need tools?From function to toolWhen one tool isn't enoughBuilding your first tool-enabled agentThe Strands Agents tool ecosystemPre-built community toolsWhat's in the toolbox?Using multiple tools togetherAWS integrationBuilding custom toolsWhen simple tools aren't enoughThe database connection problemSlow warehouse checksChoosing your approachQuick start with Strands Agent BuilderBuilding agents on LangChainSummary
Chapter 3: Agent Memory
PrerequisitesWhat is agent memoryTypes of agent memoryChallenges with implementing agent memoryUnderstanding agent workflows with memoryImplementing memory systems with AWS servicesAI agent/LLMShort-term memory (context)Structured long-term memorySemantic long-term memory (the RAG layer)Durable storageMemory orchestrationBuilding a personalized agent with memoryContext management and memory optimization for agent memoryMemory optimizationIntroducing Amazon Bedrock AgentCore MemoryArchitectural overview of Amazon Bedrock AgentCore MemoryUnderstanding Amazon Bedrock AgentCore short-term memoryMemory strategies, namespaces, and long-term retrieval in AgentCore MemoryNamespacesMemory strategiesMemory extraction moduleBringing it all togetherHands-on lab: Using AgentCore Memory with LangGraphHow the integration worksRun the agentTesting memory persistenceSummary
Chapter 4: Advanced Agent Architecture Patterns
Technical requirementsWhen one agent isn't enoughThe problem with single agentsThe multi-agent solutionWhat are multi-agent patterns?Your first multi-agent pattern: the supervisor-workerThe trade-offs of the supervisor-worker patternMake agents reusable: agent as a toolHow data flows through multi-agent systemsUnderstanding communication techniquesSequential communicationParallel communicationCollaborative communicationCommon orchestration patternsSwarm patternsGraph patternWhen multi-agent is not the answerSummary
Chapter 5: Agent Communication
Technical requirementsModel Context ProtocolWhy is a standardized protocol needed for tools?What problem does MCP solve?Understanding MCP in actionDiffernce between host and MCP clientManaging lifecycle between client and serverUnderstanding MCP server featuresToolsResourcesResource templatePrompt templatesUnderstanding MCP client featuresSamplingRootsElicitationBuilding and using MCP serversUsing pre-built MCP serversBuilding your own or custom MCP ServerHands-on example of building an MCP server and connecting to an MCP clientServer.py codeClient.py codeAgent2Agent protocolUnderstanding the difference between MCP vs A2AWhy is a standardized protocol needed for agents?Understanding A2A by an example of a multi-agent systemHow A2A architecture worksUnderstanding A2A data modelCommunication styles in A2AHow to implement A2ABuilding a multi-agent system with A2A and MCPWeather Agent ServerFlights Agent ServerTravel OrchestratorSummary
Chapter 6: Production Deployment and Enterprise Integration
Technical requirementsWhat actually changes when you go to productionDeployment pathAWS Lambda: Serverless and event-drivenAmazon ECS: Always-on and container-drivenFrom Lambda handler to web serverWhat's actually happening on AWSWhen to reach for Amazon ECSAmazon Bedrock AgentCoreThe AgentCore architectureAgentCore runtimeDeployment walkthroughHow runtime compares to Lambda and ECSAgentCore GatewayTargetsSemantic searchAuthentication: Inbound and outboundAgentCore IdentityEnterprise integration and industry use casesFinancial services: Fraud detection at transaction speedHealthcare: Multi-specialty surgical schedulingManufacturing: Predictive maintenance on the factory floorE-Commerce: Intelligent customer support with specialized agentsSummary
Chapter 7: Evaluation, Observability, and AI Governance
Technical requirementsAgent evaluationWhy agent evaluation matters?Agent evaluation use-caseRelationship between agent observability and evaluationEvaluation metrics for evaluatorsUnderstating types of agent evaluationAgent evaluation at a glanceAgentCore Evaluations providing a managed service for evaluationsCollecting traces for Agent evaluation via AgentCore ObservabilityAgentcore evaluation levelsHands-on using AgentCore evaluations to evaluate travel assistantPrerequisites:Step 1: SetupStep 2: Understand the agent (agent_app.py)Step 3: Deploy the agentStep 4: Create a custom evaluatorStep 5: Define test prompts and invoke the agentStep 6: Wait for tracesStep 7: Run evaluations (3 passes)Step 8: Summary tableStep 9: Save resultsKey insightObservability: Seeing what your agents doWhat you need to captureThe OpenTelemetry FoundationSeeing traces locally with JaegerLangfuseLangSmith: Observability from the LangChain EcosystemConnecting Strands to LangSmithWhat LangSmith and Langfuse OfferAgentCore ObservabilityWhat gets capturedExporting to external platformsPutting it all togetherObservability anti-patternsAI Governance: Keeping Agents Within BoundsGuardrails: Filtering model outputsApplying guardrails to a Strands agentAgentCore Policy: Deterministic control over tool callsHow Cedar policies workPolicies based on user identityGuardrails and policyOrganizational AI governance: The human layerAudit trailsIncident response for AI systemsBefore you go to productionSummary
Chapter 8: Unlock Your Exclusive Benefits
Unlock this book's free benefits in 3 easy steps
Index

Content preview from AI Agents on AWS

8 Unlock Your Exclusive Benefits

Your copy of this book includes the following exclusive benefits:

Follow the guide below to unlock them. The process takes only a few minutes and needs to be completed once.

Unlock this book's free benefits in 3 easy steps

Step 1

Keep your purchase invoice ready for Step 3. If you have a physical copy, scan it using your phone and save it as a PDF, JPG, or PNG.

For more help on finding your invoice, visit https://www.packtpub.com/en-us/unlock?step=1.

If you bought this book directly from Packt, no invoice is required. After Step 2, you can access your exclusive content right away.

Step 2

Scan the QR code or ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Publisher Resources

ISBN: 9781806387212

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills