Название: An Illustrated Guide to AI Agents: Concepts and Code for Building Agents with LLMs, Tools, and Memory (Final Release)
Автор: Maarten Grootendorst, Jay Alammar
Издательство: O’Reilly Media, Inc.
Год: 2026
Страниц: 448
Язык: английский
Формат: pdf, epub
Размер: 62.1 MB

Artificial Intelligence (AI) is entering a new phase. No longer limited to answering prompts or completing simple writing tasks, AI agents can now reason, plan, and act with increasing independence. From accelerating scientific breakthroughs to supporting creative work, these systems are quickly reshaping industries and everyday life. This book provides the conceptual foundation and practical insights you need to understand—and effectively work with—this emerging technology.

Through hundreds of clear graphic illustrations, Maarten Grootendorst and Jay Alammar explain how AI agents are built, how they think, and where they're heading. Designed for professionals, students, and curious learners alike, this guide goes beyond the buzz to reveal what's actually happening inside these systems, why it matters, and how to apply the knowledge in real-world contexts. With its visual storytelling and accessible explanations, An Illustrated Guide to AI Agents is your essential reference for navigating the next frontier of Artificial Intelligence.

Something remarkable happened in the years following the broad public usage of large language models (LLMs): they stopped just talking and started doing. Given a goal, a set of tools, and a bit of memory, an LLM can now search the web, write and run code, revise its own plans, and work through problems over many steps with little hand-holding. We call these systems AI agents, and they represent one of the most consequential shifts in how work gets done and how software gets built.

This book provides a comprehensive and highly visual introduction to the world of AI agents, covering both the conceptual foundations and practical applications. We start with the “brain” of today’s most powerful agents, the reasoning LLM, and then build the agent up piece by piece: memory, tools, planning, and reflection. From there we look at how agents behave as systems, how to evaluate them, and how they specialize, whether that means multiple agents collaborating, agents that can see and hear, or the coding agents that have become a staple of modern software development.

Although “illustrated” is in the name of this book, understanding agents benefits greatly from building one. Alongside the visual journey, you will build a working agent in Python, one component at a time, which we call the TinyAgent. By the end of the book, it will reason, remember, use tools, plan, and reflect. The result is essentially a package you developed yourself, with a genuine understanding of every line in it. The code printed in this book is here to teach; the code in the book’s repository is here to run.

Explore the core architecture of AI agents: tools, memory, and planning
Understand reasoning LLMs, multimodal models, and multi-agent collaboration
Learn advanced methods, including distillation, quantization, and reinforcement learning
Evaluate real-world applications, strengths, and limitations of AI agents

Prerequisites:
What you need depends on how you plan to read this book. To follow the conceptual thread, the illustrations and intuition-first explanations, you need nothing beyond curiosity: no programming, no mathematics, and no prior exposure to agent frameworks. Familiarity with LLMs helps, but Chapters 2 and 3 cover everything about them that the rest of the book relies on. To follow the hands-on thread and build the TinyAgent yourself, you will want some experience programming in Python. The deeper sections on model internals and training assume familiarity with the fundamentals of machine learning, though even there the focus is on building intuition rather than deriving equations.

Audience:
We wrote this book for several audiences at once, and we structured it so that each gets value from it:
• If you are a developer building agentic systems, the hands-on code and the TinyAgent you assemble across the chapters will give you a working foundation you understand down to the last line.
• If you come from a research or Machine Learning background, the deeper sections on model internals, training techniques, and evaluation methodology connect agent behavior back to how these models are actually made.
• If you are a technical leader, product builder, or curious reader who wants a rigorous mental model of agents without writing code, the illustrations and intuition-first explanations carry the full conceptual story on their own. The code can be skipped without losing the plot, or, fittingly, handed to your coding agent.

Ссылки удалены по требованию правообладателя

ОТСУТСТВУЕТ ССЫЛКА/ НЕ РАБОЧАЯ ССЫЛКА ЕСТЬ РЕШЕНИЕ, ПИШЕМ СЮДА!