Fajardo-HI1200×1505 99.6 KB

Andrei Fajardo

Andrei Fajardo’s new book walks you through the process of designing and implementing a complete multi-agent system from the ground up. Instead of relying on existing frameworks, you’ll learn to build the full infrastructure yourself—agents, memory modules, communication protocols, orchestration layers, and evaluation mechanisms.

In the book, you’ll explore how to:

• Build a full LLM-based agent architecture with custom data structures, loops, and tools

• Integrate with the Model Context Protocol (MCP) and extend your system to collaborate with other agents via A2A (Agent-to-Agent)

• Implement human-in-the-loop and memory-sharing mechanisms for more robust coordination

• Evaluate and iterate on your agents’ reasoning and performance in real-world tasks

By the end, you’ll have a working, extensible multi-agent system and a clear understanding of how collaborative AI ecosystems function under the hood.

Whether you’re researching agent coordination, exploring automation in AI-driven workflows, or building your own agentic infrastructure, this book offers a hands-on approach to understanding the current evolution of agent-based AI systems.

• Full details: https://www.manning.com/books/build-a-multi-agent-system-from-scratch

Don’t forget you can get 45% off with your Devtalk discount! Just use the coupon code “devtalk.com” at checkout