
Build and deploy AI agents with Python and OpenAI by exploring an end-to-end demo of agents that search, reason, use tools, and remember context to complete tasks.
Master AI agents in Python and OpenAI with a course outline, success tips, and hands-on practice, building single to multi-agent systems with memory, tools, and guardrails.
Build your first AI agent in Python using the OpenAI agents SDK, then expand to memory, tools, and a team of agents to plan a travel itinerary.
Configure OpenAI API and the agents SDK, install the OpenAI agents library, and securely manage API keys with a .env file to run AI agents using Lang chain and Pydantic.
learn to build and run your first ai agent with the openai agents sdk, mastering prompt engineering and a system prompt as a detailed job description.
Build and run an AI agent to generate a one-day Paris itinerary by configuring prompts, system instructions, and a trip planner agent with a GPT 4.1 mini model.
Experiment with the openai agents sdk by changing the itinerary request to your city and testing models like gpt-4 or gpt-5, then compare results with the upcoming solution.
Practice opportunity solution demonstrates configuring a trip planner AI agent and switching to a GPT five model, exploring city itineraries like Cairo with authentic food and museum stops.
Test agent memory by running a trip planner and verify it forgets prior conversations unless memory is configured, while exploring traces on the OpenAI platform, token usage, and tokenization.
Develop a warm, calming mental health coach agent that generates short self-care tips with emojis, avoids medical diagnostic advice, using the OpenAI agents SDK and GPT five mini, with traces.
Define a warm mental health coach agent with a system prompt using the OpenAI agents SDK, delivering a concise mood tip under 200 characters and an emoji.
Build an AI agent for personal finance in Python and add memory with a SQLite session to recall past conversations using the OpenAI API and the OpenAI agents SDK.
Set up the environment and OpenAI API key from a .env file, configure a Python agent with the OpenAI SDK, and test a stateless finance advisor with a runner.
Equip your AI agent with memory using a SQLite session to store conversation history, enabling recall of past prompts and outputs in a finance advisor scenario.
Build an ai wellness coach agent with memory using sqlite session and the OpenAI sdk to propose one small weekly health habit and remember conversations, while avoiding repeated habits.
Learn to create a memory-enabled wellness coach AI agent in Python, using SQLite session memory to remember prompts and suggest one simple weekly health habit.
Set up OpenAI and Tivoli API keys, define a Tivoli search function as a tool, and build an AI agent that uses real-time search and a code interpreter.
Set up the OpenAI and Tavileh web access APIs, load your API keys into the environment file, and enable your agent to perform real-time web searches.
Define a typed dictionary based search function and expose it as a tool for your AI agent with the function tool decorator, enabling queries and results from Tavileh API.
Modify the search function to return only the titles of results, excluding the content. Increase the max results parameter from two to three by updating the code.
Update the practice opportunity solution by configuring a search function and decorator to default to three results and return only titles.
Build and run an AI agent with a real-time web search tool, add memory with SQLite sessions, configure the agent as a market researcher, and trace tool usage for debugging.
Use the live researcher agent to find and summarize the key features of your favorite vehicle, test its memory (sqlite session) and search capability with a follow-up question.
Learn to run ai agents with tools using a live researcher agent and runner, executing session-based questions and summarizing results with openai built-in tools, demonstrated via a Tesla Cybertruck example.
Leverage built-in OpenAI tools: web search, file search, vector stores with Rag, and the code interpreter to run Python calculations and real-time data analysis.
Build a team of AI agents that coordinate to achieve a goal, featuring a researcher, analyst, and writer, plus a manager function to orchestrate the workflow using the OpenAI API.
Install the OpenAI agent toolkit and configure api keys, import libraries, and expose a search tool via a decorator to fetch titles and content for up to three results.
Configure two ai agents—the researcher and the analyst—using a search tool to produce concise research summaries and an analysis of trends, risks, and insights.
Formulate a research question on electric vehicle and battery trends, run the researcher agent for findings, then pass the summary to the analyst agent to extract trends, risks, and insights.
Learn to run a researcher agent to summarize electric vehicle battery data and pass it to an analyst agent for a concise EV market analysis, including lithium iron phosphate.
Configure the third agent, a writer, to produce an executive summary, a markdown report of 500+ words, and 3–5 follow-up questions, using the pedantic library for data validation.
Rewrite the writer's agent instructions so outputs are generated entirely in French, and test a concise French report in the next task. Build a manager function to orchestrate these steps.
Explore how to reinforce agent instructions, change outputs to French, add emphasis, and ground results in attached documents to improve model performance in a practice opportunity solution.
Orchestrate a full ai pipeline by running researcher, analyst, and writer agents to produce a complete report, using a sqlite session memory and an asynchronous manager.
Build a multi-agent pipeline using the writer agent in French, then deploy a creative director, strategist, and copywriter to brainstorm 3–5 ideas, select options, and craft tweets for a campaign.
Explore ai agents for beginners by building a multi-agent team in Python and OpenAI to generate creative advertising campaigns, with a creative director, strategist, and tweet copy in a pipeline.
Design AI agents with guardrails and handoffs, coordinating planner, writer, search, and fundamental analysis tools, plus memory and orchestration via sessions, to generate an executive-ready report.
Set up api keys and tools for ai agents, install OpenAI agents 0.2.2, and configure main and nano models; define a search tool with guardrails.
Learn how to define a planar AI agent with built-in guardrails, implement a politics guardrail that detects political topics, and test a planner agent that generates multi-query search plans.
Design and test a defense guardrail for AI agents, teaching how to detect sensitive inputs like defense topics and trigger appropriate safety responses.
Develop and test a defense guardrail for AI agents that blocks questions about defense topics, elections, and government policy, returning a reasoning so users understand why the guardrail triggers.
Define a search agent and a fundamental analysis agent as tools to be used by an AI agent, enabling web search for information and analysis of revenues, debts, and ratios.
Learn how to build and connect AI agents as tools, including writer, search, and fundamentals agents, to generate comprehensive reports with executive summaries and follow-up questions.
Review the traces log to identify tools used and their frequency, explain the trace hierarchy and workflow, and perform a sanity check by inspecting OpenAI API traces after logging in.
Explore monitoring agent traces and hierarchy in a multi-agent workflow, observing execution time as the writer agent orchestrates searcher and fundamentals analysts.
Explore handoffs between planner and writer agents using the OpenAI agents SDK, cloning the planner and passing user queries and plans to the writer, with guardrails ensuring handoffs and reporting.
Create a sentiment analysis agent named sentiment agent that searches for a company and analyzes online sentiment as positive, negative, or neutral, to be used alongside existing agents and reviewed.
Learn to extend an AI agent with a sentiment analysis tool, empowering a writer agent to produce executive summaries and market sentiment insights using search and fundamentals.
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.