Data For Science

Agentic Design Patterns: A Hands-On Guide to Building Intelligent Systems

by A. Gullí

Reviewed in Data Science Briefing #310,

A. Gullí's "Agentic Design Patterns: A Hands-On Guide to Building Intelligent Systems" feels like a timely guide for data scientists and machine learning engineers who are ready to move past the hype around AI agents and actually build them. The book covers foundational agentic patterns, like reflection, tool use, planning, and multi-agent collaboration, in enough detail to be useful without getting lost in abstract theory. The practical code examples help ground the concepts, and the emphasis on system-level thinking, what to build, how to evaluate it, and where things break, makes it genuinely relevant for production-oriented teams.

What stands out is the book's focus on patterns, not just tools. It treats agents as design problems, not feature lists, and encourages the reader to think carefully about when to delegate to an LLM and when to keep humans in the loop. For ML engineers already building pipelines and serving models, this adds a useful layer: how to structure the reasoning, memory, and coordination that sit on top of inference.

The book is not without limits. Some sections feel rushed, and the coverage of evaluation, monitoring, and failure modes could go deeper, especially for teams working at scale. But as a first pass at organizing the agentic space into something practical and composable, it does the job well. If you're building with LLMs and want a framework for thinking about agents, not just a recipe, this is a solid starting point.

Read another review

All 72 book recommendations

​

Subscribe to get our latest content by email.
    We won't send you spam. Unsubscribe at any time.