Skip to Content
View all events

Design Patterns for AI Agents

Published by O'Reilly Media, Inc.

Intermediate content levelIntermediate

Build systems that are adaptive, scalable, and collaborative

What you’ll learn and how you can apply it

  • Design intelligent agents by applying reusable patterns for perception, reasoning, action, and learning
  • Compare and select agent strategies based on application needs
  • Structure multi-agent systems using interaction, coordination, and architectural patterns for scalability and collaboration
  • Integrate advanced capabilities such as planning, memory, autonomy, and human-agent interaction into practical agent designs

Course description

This course with Shaun Wassel explores how to apply reusable design patterns to AI agents, enabling a structured way for practitioners to build systems that are adaptive, scalable, and collaborative. By framing agent development in terms of proven patterns (much like software engineering) you’ll gain a practical tool kit that accelerates design and avoids reinventing the wheel. As AI agents increasingly move from labs to real-world applications, understanding these patterns is essential for creating agents that can think, learn, and work effectively with humans and each other.

This live event is for you because...

  • You’re an AI practitioner, engineer, or researcher looking to structure agent systems with proven reusable patterns.
  • You want practical design tools that go beyond theory to help you build intelligent agents faster and more effectively.
  • You need to bridge gaps between software design principles and modern AI agent architectures.
  • You want to create agents that scale and collaborate with other agents or directly with humans in real-world applications.

Prerequisites

  • A computer with internet access for viewing materials
  • (Optional) Coding examples can be followed using Python in any standard environment (e.g., Jupyter Notebook, Google Colab, or a local IDE)
  • Basic knowledge of AI and machine learning concepts, plus familiarity with software design principles or object-oriented programming

Recommended preparation:

Recommended follow-up:

Schedule

The time frames are only estimates and may vary according to how the class is progressing.

Introduction and setup (65 minutes)

  • What are AI agents?
  • Overview of GitHub Copilot
  • Setting up Python and Copilot
  • Development environment basics
  • Q&A
  • Break

Core concepts of AI agents (60 minutes)

  • Agent architecture basics
  • Working with APIs and tools
  • Handling inputs and outputs
  • Prompting strategies in code
  • Break
  • Q&A

Building agents with Python and Copilot (60 minutes)

  • Writing agent logic with Python
  • Using Copilot for rapid prototyping
  • Managing state and memory
  • Debugging and testing agents
  • Break
  • Q&A

Advanced strategies and applications (60 minutes)

  • Multistep reasoning and workflows
  • Integrating external libraries and APIs
  • Scaling and performance considerations
  • Practical use cases and examples
  • Q&A

Your Instructor

Shaun Wassell

Shaun Wassell is a lifelong programmer and problem-solving addict. His goal is to help people build incredible software and solve meaningful problems by mastering the art of software development. For the past 2+ years, he's been a trainer at CBT Nuggets, and focuses on creating high-quality web development and certification content. You can check out his extensive collection of React, Angular, Python and JavaScript courses, as well as a huge amount of other software- and IT-training content at cbtnuggets.com.

linkedinXsearch

Skills covered

  • Design Patterns
  • Generative AI
  • Artificial Intelligence (AI)