Building Agentic AI Systems: A strategic guide to designing and governing autonomous AI systems (English Edition) - Tapa blanda

Sinopsis

The field of artificial intelligence is undergoing a paradigm shift, moving beyond instruction-based generative AI to intent-driven agentic AI. This new frontier is defined by autonomous systems that can independently reason, plan, and act to achieve complex goals, serving as a strategic imperative that promises to redefine operational efficiency for the modern enterprise.

This book provides a hands-on playbook for architects and technology leaders to move from prototypes to robust, enterprise-grade agentic applications. It guides you through the lifecycle of building autonomous systems, mastering durable patterns like Coordinator-Worker-Delegator (CWD), building for interoperability using the open-standard Agent Communication Protocol (ACP), and mitigating security threats with a formal Agentic Risk Framework.

By the end of this book, you will be equipped with the expert knowledge required to architect, build, deploy, and govern sophisticated, scalable, and value-driven agentic AI systems that deliver tangible results in real-world enterprise environments.

What you will learn

● Architect autonomous AI agents with modern Python frameworks.

● Implement multi-agent workflows using LangGraph and AutoGen.

● Integrate the Model Context Protocol for API tool utilization.

● Manage state and memory in long-running agentic systems.

● Deploy dynamic retrieval-augmented generation (RAG 2.0).

● Establish rigorous security protocols and ethical AI guardrails.

● Scale enterprise AI systems for complex production environments.

Who this book is for

This book is for AI developers, machine learning engineers, software architects, technology leaders, and product managers aiming to build the next generation of autonomous enterprise systems. Readers should have an intermediate understanding of Python and foundational concepts in large language models.

Table of Contents

1. From Instruction to Intent

2. Anatomy of an Intelligent Agent

3. Principles of Multi-agent Collaboration

4. Coordinator Worker Delegator Pattern

5. Architecting for Interoperability with the Agent Communication Protocol

6. Modern Agentic Stack and its Decision Framework

7. Evaluating Agent Performance with Business Value Metrics

8. Observability and Debugging for Autonomous Systems

9. Economics of Agency Managing Cost and ROI

10. Architecting Trust with the Agentic Risk Framework

11. Designing for Collaboration with Human-in-the-Loop Patterns

12. Moral Compass of Ethical Frameworks and Accountability

13. Sector Deep Dives in Agentic AI Production

14. Next Frontier in Architecting Self-improving Systems

"Sinopsis" puede pertenecer a otra edición de este libro.