Build a custom multi-agent AI framework from scratch and orchestrate collaborative agent workflows using MCP and A2A. Key Features Create a fully functional agentic AI system without relying on external frameworks Extend agent capabilities using MCP-based context and A2A collaboration Build, test, and deploy real-world AI agents that reason, plan, and act Book DescriptionTired of black-box AI agent frameworks? This book gives you full control by teaching you how to design and implement your own extensible, production-grade agentic AI framework in Python. You'll start by mastering the foundational concepts behind generative AI and intelligent agents before progressing to practical implementation - beginning with a simple tool-using agent and advancing to multi-agent orchestration using Model Context Protocol (MCP) and Agent-to-Agent (A2A) messaging. Through annotated code walkthroughs, real-world engineering patterns, and deployment scenarios, you'll build everything from custom tool interfaces to collaborative agent workflows for distributed environments. You'll explore how agents sense, plan, act, use tools, manage context, and coordinate with other agents to achieve complex goals. Written by Gigi Sayfan, an experienced engineer and bestselling author, this book empowers you to deeply understand agentic architectures and build systems that are adaptable, transparent, and powerful. Whether you're building AI products with LLMs or experimenting with advanced orchestration, this book gives you the skills to go beyond prebuilt solutions and engineer your ownWhat you will learn Design and implement tool-using AI agents from the ground up Build modular components for extensible agent frameworks Create secure and observable tools with structured inputs Integrate agents with chat UIs like Slack and Chainlit Leverage MCP for context handling and agent memory Orchestrate collaborative agent workflows using A2A Debug and deploy agents in production-like environments Explore future-ready agent capabilities and GenUX design Who this book is forThis book is essential for AI engineers, ML practitioners, and software architects building agentic systems with large language models. It's also ideal for DevOps engineers and technical leaders seeking deep insights into building and scaling autonomous AI workflows. Python coding skills and basic familiarity with LLMs are recommended.

About the Author
Gigi Sayfan has been developing software for 25+ years in domains as diverse as instant messaging, morphing, chip fabrication process control, embedded multimedia applications for game consoles, brain-inspired ML, custom browser development, web services for 3D distributed game platforms, IoT sensors, virtual reality, and genomics. He has written production code in languages such as Go, Python, C, C++, C#, Java, Delphi, JavaScript, and even Cobol and PowerBuilder for operating systems such as Windows (3.11 through 7), Linux, macOS, Lynx (embedded), and Sony PlayStation. His technical expertise includes databases, low-level networking, distributed systems, containers, unorthodox user interfaces, modern web applications, and general SDLC.


Book Information
ISBN 9781806116478
Author Gigi Sayfan
Format Paperback
Imprint Packt Publishing Limited
Publisher Packt Publishing Limited