ManningBooks

ManningBooks

Devtalk Sponsor

Agent Design Patterns (Manning)

AI agents are showing up everywhere—and most of them are being built by trial and error. Agent Design Patterns distills the experience of a growing community of agent builders into 20+ composable, reusable patterns for building AI agents that are reliable, efficient, controllable, and easy to reason about. In this uniquely valuable book, author and NVIDIA Research scientist Peter Belcak gives you a durable engineering vocabulary and design intuition that outlasts any new model release.

Peter Belcak

Agent Design Patterns by Peter Belcak is about moving beyond one-off prompts and treating agents as systems that can be designed, tested, and improved. It introduces reusable patterns for common challenges in agent development, particularly around capability, reliability, and robustness.

The book takes a practical approach. Peter builds a mortgage advisor to demonstrate how the patterns work together, then explores individual patterns such as:

• Errand, for breaking complex workflows into manageable units
• Guardrail, for gating actions and improving reliability
• Resampler, for using repeated sampling to produce more dependable results

Five chapters are currently available through Manning’s Early Access Program. Readers can follow the book as Peter writes it, receive new chapters when they’re released, and send feedback that may help shape the finished book.

If you’re already experimenting with agents, I’d be interested to hear which design problems have caused you the most trouble. For me, the gap between an impressive demo and an agent you can actually depend on is where this topic gets particularly useful.


Don’t forget you can get 45% off with your Devtalk discount! Just use the coupon code “devtalk.com” at checkout :+1:

First Post!

mira_tokarev

mira_tokarev

The distinction between making an agent capable and making it dependable feels important. Guardrails help with the obvious unsafe actions, but recovery after a partial tool failure is where many systems still become brittle. I’d be interested in seeing the patterns tested under interrupted runs: can the agent resume from a checkpoint without repeating side effects, and can it explain why it chose a fallback? Resampling seems useful for ambiguous decisions, but I’d probably reserve it for the steps where uncertainty is measurable rather than making every workflow more expensive.

Where Next?

Popular Ai topics Top

New
ManningBooks
Rearchitecting LLMs: Structural techniques for efficient models turns research from the latest AI papers into production-ready practices ...
New
ManningBooks
Hugging Face in Action reveals how to get the absolute best out of everything Hugging Face, from accessing state-of-the-art models to bui...
New
ManningBooks
In Designing AI Agents, you’ll learn how to establish agent architectures that manage costs and take governance seriously from day one. T...
New
ManningBooks
Context engineering is the discipline of selecting, organizing, updating, compressing, prioritizing the precise context a model needs to ...
New
smarties-press
Bertrand Meyer The most important technology of our time — decoded Millions use AI every day. Few understand why it works. ✓ Why large ...
New
ManningBooks
General-purpose models are impressive, but they aren’t always the right fit for domain-specific or mission-critical applications. Prompt ...
New
ManningBooks
Until recently, writing GPU kernels for LLM training and inference meant learning low-level programming tools like CUDA and C++. Triton, ...
New
ManningBooks
AI is everywhere—from the tools we use at work to the services and products we rely on every day. But keeping up with the terminology, ca...
New
ManningBooks
Customers expect fast, accurate suggestions that are perfectly aligned to their preferences, even when they have limited historical data ...
New

Other popular topics Top

PragmaticBookshelf
Instantly view the changes you make to an app with stateful hot reload, and define a declarative UI in the same language as the app logic...
New
dasdom
No chair. I have a standing desk. This post was split into a dedicated thread from our thread about chairs :slight_smile:
New
PragmaticBookshelf
Design and develop sophisticated 2D games that are as much fun to make as they are to play. From particle effects and pathfinding to soci...
New
Rainer
My first contact with Erlang was about 2 years ago when I used RabbitMQ, which is written in Erlang, for my job. This made me curious and...
New
PragmaticBookshelf
Build highly interactive applications without ever leaving Elixir, the way the experts do. Let LiveView take care of performance, scalabi...
New
PragmaticBookshelf
Author Spotlight Jamis Buck @jamis This month, we have the pleasure of spotlighting author Jamis Buck, who has written Mazes for Prog...
New
New
PragmaticBookshelf
25 puzzles that will make you a better C programmer by challenging your knowledge of the language, and explaining the technical details o...
New
Fl4m3Ph03n1x
Background Lately I am in a quest to find a good quality TTS ai generation tool to run locally in order to create audio for some videos I...
New
pragdave
Build robust LLM-powered apps, chatbots, and agents while mastering AI engineering principles that will help you outlast the tools and th...
New