Building Advanced Agentic Systems on AWS

In this course, you’ll learn about implementing production-ready multi-agent systems using Amazon Bedrock AgentCore, covering mulit-agent patterns, context optimization techniques, security configurations, and monitoring frameworks. You will examine the skills needed to move beyond proof-of-concept to scalable, secure, and observable agentic AI implementations. The course prepares you to design and deploy advanced agentic systems ready for real-world production environment.

Course objectives

This course includes presentations, hands-on lab, and group exercises.

In this course, learners will:

Analyze scenarios that require multi-agent architectures.
Describe primary multi-agent communication patterns and their use cases.
Configure agent-as-tool patterns for production deployments.
Implement memory sharing using available platform capabilities.
Implement context management strategies for production workloads.
Design context compression and optimization mechanisms.
Optimize resource usage and cost management across multi-agent systems.
Configure policy-based access control using AgentCore Policy Engine.
Implement VPC integration for secure agent deployments.
Implement distributed tracing and monitoring across multi-agent systems.
Establish comprehensive agent evaluation frameworks.
Configure integration patterns for enterprise observability systems
Establish comprehensive audit trails and compliance monitoring.
Integrate agentic systems with production APIs and services.
Design deployment strategies for production environments.
Assess production readiness and establish continuous improvement processes.

Target Audience

Course level: Advanced (300)
Software developers seeking intermediate knowledge for building advanced agentic AI systems
Technical professionals exploring AI capabilities and interested in building advanced agentic AI systems.
Development teams building advanced agentic AI solutions.

Prerequisites

We recommend that attendees of this course have:

Agentic AI Foundations
Building Agentic AI with Amazon Bedrock AgentCore

Course content
FAQ

Module 1: Multi-Agent Architecture and Communications Patterns

Single agent limitations and multi-agent benefits
Instructor Demonstration: Building and Deploying Intelligent Financial Agents with Amazon Bedrock Strands and AgentCore - Task 1: Building a Personal Budget Assistant with Strands Agents
Multi-agent communication patterns
Memory and state management
Instructor Demonstration: Building and Deploying Intelligent Financial Agents with Amazon Bedrock Strands and AgentCore - Task 2: Building a Multi-Agent System for Complex Financial Analysis

Instructor Demonstration: Building and Deploying Intelligent Financial Agents with Amazon Bedrock Strands and AgentCore

Module 2: Context Engineering and Performance Optimization

Context as finite resource
Context optimization techniques
Tool design for context efficiency

Module 3: Security and Compliance Implementation

Production Identity and Access Management
VPC integration and network security

Module 4: Production Monitoring, Observability, and Evaluation

Monitoring architecture
AgentCore evaluation
Enterprise observability integration
Instructor Demonstration: Building and Deploying Intelligent Financial Agents with Amazon Bedrock Strands and AgentCore - Task 3: Deploying Production-Ready Agents with Amazon Bedrock AgentCore

Module 5: Well-Architected Agentic AI Systems

Applying the Well-Architected framework
Well-Architected deployment
Production readiness

Module 6: Course Wrap-up

Next steps and additional resources
Course summary

Practical information

Duration: 1 day
Price: 9.900
Language: English, Norwegian
Format: Can be delivered as an open course or as an in-house course

FAQ

Hva betyr “agentic systems”?
Agentic systems er intelligente systemer som kan oppfatte mål, planlegge handlinger og utføre oppgaver på egen hånd ved hjelp av AI-agentteknikker, ofte med tilgang til verktøy og API-er for å løse komplekse problemer.

Hva er forskjellen mellom dette kurset og grunnleggende agentic AI-kurs?
Dette kurset går dypere inn i avanserte teknikker for å bygge robuste, skalerbare agentbaserte løsninger på AWS, inkludert integrasjon med flere tjenester, forbedret sikkerhet, orchestrering og produksjonsklar drift.

Hvilke AWS-tjenester er relevante for kurset?
Kurset dekker typiske AWS-tjenester som Amazon Bedrock, Lambda, EventBridge, Step Functions, DynamoDB, API Gateway og andre komponenter som er relevante for å drifte og orkestrere avanserte agentiske systemer.

Trenger jeg erfaring med AWS for å ta dette kurset?
Ja, det anbefales at du har grunnleggende kjennskap til AWS, særlig IAM, Lambda og serverløse arkitekturer, for å få fullt utbytte av kurset.

Hvordan brukes agentic AI i praksis?
Agentic AI brukes til å automatisere oppgaver som krever beslutningstaking, arbeidsflytkoordinering, flertrinnsinteraksjon med data- og tjenestesystemer, og kan forbedre effektivitet og pålitelighet i løsninger som kundedialog, datarensing, analyse og intelligent automatisering.

Kursavgift:	9,900 NOK
Kurset inkluderer:	Course documentation and exercises. Lunch and refreshments for in class events only.
Timer	09:00 - 17:00
Varighet:	1 dager
Språk	Norwegian, English