UdemySoftware DevelopmentPaid courseAll LevelsCertificate

AI Agents for Beginners: Build Agents in Python & OpenAI (Udemy.com)

Build Real AI Agents with Tools, Memory, Handoffs, & Guardrails Using OpenAI Agents SDK & Python.

4.5

194 reviews

Created by: Prof. Ryan Ahmed, PhD, MBA

Last updated December 2025

What you will learn

Build AI agents in Python using OpenAI Agents SDK with Memory and Tools.
Build a travel planner AI Agent that creates an itinerary based on user preferences.
Inspect the agent’s behavior using traces on the OpenAI API platform.
Master best practices in creating Autonomous AI Agents
Master the art of Prompt Engineering & Setting Up Agent Instructions
Understand the Key components of an Effective AI Agent System Prompt (Context, Instruction, Input, & Output Indicator)
Test AI agents and watch it run using the Runner class
Design & build a personal finance AI agent with memory that can recall user preferences, past tasks, & ongoing conversations to deliver personalized experiences
Integrate SQLite-based session storage to enable persistent memory for agents
Test and compare agent behavior before and after adding memory to evaluate improvements in context retention.

Course Description

In this course, you’ll learn how to build real AI agents in Python using the OpenAI Agents SDK. You’ll understand the core building blocks of modern AI agent systems, including reasoning, memory, tool usage, planning, and guardrails and how these components work together to automate and enhance tasks.

You will also learn how to develop a team of autonomous AI Agents that can work together to achieve a goal. You’ll learn how to inspect and debug agent behaviour using tracing and observability tools, so your agents are not just powerful, but also transparent, testable, and safe.

By the end of this course, you’ll be able to:

Build AI agents in Python using the OpenAI Agents SDK
Understand how agents reason, plan, and execute tasks using system prompts and instructions
Design effective agent system prompts (context, instructions, inputs, and outputs)
Understand how AI agents work under the hood
Use tools, memory, and planning to create more capable, context-aware agents
Design safe agent interactions with guardrails and controlled execution
Compare agent behaviour with and without memory to evaluate context retention

This course is designed for developers and engineers who want a practical, code-first introduction to AI agents. No prior experience with agent architectures or machine learning is required just basic Python knowledge and an interest in building real AI systems that work reliably in practice.

All Levels

Time: 4.5h

Udemy Discount

The discount is applied through our link. Open the course from here and Udemy's current promotional price is applied at checkout on most courses, no code to type.

Get the Udemy discount

Some courses are excluded from Udemy's promotions. If the price does not drop, clear your browser cookies and use the button again.

Instructor Details

Prof. Ryan Ahmed, PhD, MBA

4.5 Rating
194 Reviews

I'm Dr. Ryan Ahmed, professor, engineer, and founder of Stemplicity, where we help people get past the hype and actually build things with AI, Agentic AI, Cloud, and Data Science.

Over the past 10 years, I've taught 1 million learners across 160 countries — 700,000+ enrolled in my Udemy courses, 260,000+ subscribers on the "Prof. Ryan Ahmed" YouTube channel, and 120,000+ on Coursera. I also run corporate AI training for teams at HSBC, RBC, Discover, and Barclays across the US, Canada, and the UK. I held leadership roles at GM, Samsung, and Stellantis in Canada and the U.S., working on electric and autonomous vehicle technologies.

I hold a MASc, PhD, and MBA from McMaster University. I’m also a licensed Professional Engineer and a Stanford-certified program manager with over 60+ published research papers in AI and battery systems.

But credentials are the least interesting part. Here's what I actually believe: "AI is the biggest opportunity of our lifetime", and most people are sitting it out because they think they need to code or have a PhD. You don't. If you're willing to show up and try, I'll help you get good at this.