Chapter 1: Understanding Multi-Agent Systems

Introduces agents that reason, act, communicate, and adapt. Defines multi-agent systems for complex tasks requiring planning, diverse expertise, and adaptive solutions. Build your first poet-critic collaboration using round-robin orchestration.

Chapter 2: Multi-Agent Patterns

Taxonomy of orchestration patterns from explicit to emergent control. Deterministic workflows (sequential, parallel, supervisor) offer predictable control. Autonomous patterns (plan-based, handoff, conversation-driven) provide flexibility with trade-offs.

Chapter 3: UX Principles for Multi-Agent Systems

Shift from direct manipulation to delegation design with natural language. Four design principles: capability discovery, cost-aware action delegation, observability and provenance, and interruptibility for autonomous workflows.

Chapter 4: Building Your First Agent

Implement core agent loop with async architecture, streaming, and cancellation. Build picoagents progressively from Q&A agents to multi-tool workflows. Agent abstraction with multiple LLM providers and memory interfaces.

Chapter 5: Building Computer Use Agents

Extend framework for interface automation in browsers and applications. Three components: action generation, interface representation (DOM/visual/hybrid), and execution. Implicit vs explicit planning and hierarchical composition.

Chapter 6: Building Multi-Agent Workflows

Workflows as computational graphs with nodes as computation units and edges as control flow. Type-safe steps with Pydantic, conditional edges, dependency resolution, concurrent execution, and automatic checkpointing.

Chapter 7: Building Autonomous Multi-Agent Orchestration

Orchestrator loop: select agent, execute turn, check termination, repeat. Composable termination conditions and round-robin orchestration. Foundation for AI-driven selection and plan-based orchestration.

Chapter 8: Building Modern Web Experiences

Build web interfaces for long-running agent tasks with real-time progress. Two-component architecture: backend (agent execution) and frontend (UI). Complete app in ~200 lines using FastAPI + SSE and JavaScript.

Chapter 9: Multi-Agent Frameworks

Survey AutoGen, LangGraph, CrewAI through pattern lens. Compare architectures, orchestration approaches, and design philosophies. When to build from scratch vs leverage existing frameworks.

Chapter 10: Evaluating Multi-Agent Systems

Trajectories as fundamental evaluation unit. Build framework with metrics and judges. Reference-based evaluation, LLM judges, and practical planning strategies for measuring and improving systems.

Chapter 11: Optimizing Multi-Agent Systems

Ten common failure modes with optimization strategies. Making small models effective through optimization techniques. Automatic optimization for systematically improving parameters without manual tuning.

Chapter 12: Protocols for Distributed Agents

Distributed architectures across machines/organizations. Agent requirements: streaming, resumability, durable state. Model Context Protocol (MCP) for tools, Agent-to-Agent (A2A) for collaboration.

Chapter 13: Ethics and Responsible AI

Four dimensions: controllability, action capability, domain scope, verification. Agentic noise as platform imbalance. Distributed responsibility, emergent risks, security as ethical imperative. Behavioral alignment checklist.

Chapter 14: Answering Business Questions from Unstructured Data

Analyze thousands of YCombinator companies to identify AI agent builders. Four-stage workflow: data loading, pre-filtering, structured LLM analysis, insight generation. Two-stage filtering, checkpointing, cost monitoring.

Chapter 15: Software Engineering Agent

Build coding assistant like Copilot/Cursor. Agent capability from tools + prompts + memory. Five tool categories, detailed prompts encoding workflows. Complete calculator module demonstrating LLM-to-specialist transformation.