Overview
This tutorial explores OpenAI's Agents SDK, a Python-first agent development library comparable to LangChain, Llama-Index, and Pydantic AI. Learn about the key features including the automated agent loop for tool calls and LLM interactions, Python-first approach that leverages existing Python features without new abstractions, guardrails with parallel input validations, and function tools that convert Python functions into tools with automatic schema generation. Follow along with practical code examples as the instructor demonstrates how to implement agents, runners, function tools, guardrails, and conversational agents. The tutorial includes access to a GitHub repository with code examples and timestamps for easy navigation through different sections covering the essentials of this new SDK.
Syllabus
00:00 OpenAI Agents SDK
01:05 Agents SDK Code
02:38 Agent and Runner
06:56 Function Tools
12:13 Agents SDK Guardrails
18:29 Conversational Agents
21:02 Thoughts on Agents SDK
Taught by
James Briggs