OpenAI Agents API: Build and Run Managed AI Agents
Last updated 11/2026
Created by Sergey Kargopolov
MP4 |
Video: h264, 1920x1080 |
Audio: AAC, 44.1 KHz, 2 Ch
Level: All Levels |
Genre: eLearning |
Language: English + subtitle |
Duration: 25 Lectures ( 2h 0m ) |
Size: 985.6 MB
Learn to create OpenAI agents, configure tools and environments, and build your first agent application using Codex
What you'll learn

Understand how the OpenAI Agents API works and how managed AI agents differ from traditional AI model requests

Create and configure AI agent definitions using the OpenAI Platform

Create OpenAI-hosted execution environments and configure packages, setup commands, and network access

Configure OpenAI Vault secrets to securely provide credentials for accessing private GitHub repositories

Understand how agents use custom functions, sessions, and event streaming to perform multi-step tasks

Use OpenAI Codex to build and run an application that communicates with the Agents API and investigates Java projects for security vulnerabilities
Requirements

Basic understanding of software development and APIs is helpful, but not required

No prior experience with AI agents or the OpenAI Agents API is needed

An OpenAI Platform account with API access and available credits

A computer with internet access and a web browser

No advanced programming skills are required. We will use OpenAI Codex to help build our sample application
Description
Learn how to build and run managed AI agents using the
OpenAI Agents API.
This beginner-friendly course explains how managed AI agents work and shows you how to create and run one using the OpenAI Platform.
What You Will Learn

What AI agents are and how they differ from simple model requests.

How the AI model and OpenAI-managed Codex harness work together.

How to create and configure agents, tools, and execution environments.

How to manage GitHub credentials using OpenAI Vault.

How sessions, events, items, and custom function calls work.

How applications communicate with the Agents API using HTTP endpoints.

How to use OpenAI Codex to build an application that connects to your agent.
Hands-On Project
As a practical project, you will create a
Java Security Investigator agent. It will examine a Java project, research security advisories, and prepare an investigation report.
You will also configure a send_email function. For this course, the function prints the email details to the terminal instead of sending an actual email.
You will use
Codex to generate the application code, so you can focus on understanding the Agents API rather than writing every HTTP request yourself.
No previous experience with AI agents is required. This course is designed for developers who want a practical introduction to building applications with OpenAI-managed agents.
Who this course is for

Software developers who want to learn how to build and run managed AI agents using the OpenAI Agents API

Developers interested in adding AI agent capabilities to their existing applications

AI enthusiasts who want to understand how managed AI agents work and how they use tools, sessions, and execution environments

Developers who want to learn how to use OpenAI Codex to build applications that communicate with AI agents

Beginners who are new to the OpenAI Agents API and want a practical, step-by-step introduction
Homepage
Code:
https://www.udemy.com/course/openai-agents-api-build-and-run-managed-ai-agents