Agentic architectural patterns for building multi-agent systems : proven design patterns & practices for GenAI, Agents, RAG, LLMOps & enterprise-scale AI systems / Dr. Ali Arsanjani, Juan Pablo Bustos.

Format:

Book

Author/Creator:

Arsanjani, Ali, author.

Bustos, Juan Pablo, author.

Language:

English

Subjects (All):

Artificial intelligence.

Intelligent agents (Computer software).

Physical Description:

1 online resource

Place of Publication:

Birmingham : Packt Publishing, 2026.

Summary:

Transform GenAI experiments into production-ready intelligent agents with scalable AI systems, architectural patterns, frameworks, and responsible AI and governance best practicesFree with your book: DRM-free PDF version + access to Packt's next-gen Reader* Key Features Build robust single and multi-agent GenAI systems for enterprise use...

Contents:

Intro

Agentic Architectural Patterns for Building Multi-Agent Systems

Proven design patterns and practices for GenAI, agents, RAG, LLMOps, and enterprise-scale AI systems

Foreword

Contributors

About the authors

About the reviewers

Join our Discord and Reddit space

Table of Contents

Preface

Who this book is for

What this book covers

To get the most out of this book

Download the example code files

Download the color images

Conventions used

Get in touch

Free Benefits with Your Book

How to Unlock

Share your thoughts

Part 1

Foundations and Core Agent Concepts

1

GenAI in the Enterprise: Landscape, Maturity, and Agent Focus

The transformative potential of GenAI

Overview of business applications

Horizontal applications (cross-functional use cases)

Vertical or domain-specific applications

Introducing agentic AI systems

The anatomy of agentic AI

Core components

Agent anatomy

Data stores and environment context

Key architectural features

The GenAI Maturity Model: a path to agentic systems

The new agentic stack

Enabling agent communication: from tools to collaboration

Agent internals (common to A and B for simplicity)

The MCP server

The agent server

Challenges hindering production-grade GenAI

Summary

Get This Book's PDF Version and Exclusive Extras

2

Agent-Ready LLMs: Selection, Deployment, and Adaptation

Role of LLMs in agentic systems

Model selection: choosing the right foundation

Context window size

Model size and specialization for agents

Native support for tool use and function calling

Model robustness, reliability, and safety

Adaptability and fine-tuning potential

Other key selection considerations

Deployment and performance optimization for agents

Serving architectures for agentic LLMs

Cloud-hosted APIs

Self-hosted models

Edge deployment

Performance optimization strategies

Latency reduction

Throughput maximization

Cost optimization

Optimizing for tool interaction

Security considerations in LLM deployment for agents

AgentOps: managing LLMs in agentic systems

Subscribe for a free eBook

3

The Spectrum of LLM Adaptation for Agents: RAG to Fine-tuning

From generic LLMs to specialized agents

Another maturity model for agentic AI

The granularity of agents

A hierarchical agentic architecture for business process automation

The core components: agents and their capabilities

The hierarchical structure: orchestrators and specialists

Governance and observability via callbacks

Contextual enhancement: stage 1, enhancement with RAG

RAG-powered customer support agent

Financial analyst agent leveraging RAG for timely market insights

Compliance agent ensuring adherence with RAG in transaction monitoring

Analyzing the scenarios

Fine-tuning for agentic capabilities

Domain specialization

Notes:

OCLC-licensed vendor bibliographic record.

ISBN:

1-80602-957-X

OCLC:

1559037681

Publisher Number:

CIPO000313488