Project

Butler Agent

Easy

42 completions

~ 1 hour

4.5

Learn to utilize OpenAI's function calling, the Think-Act-Observe loop pattern, and the Responses API to build an autonomous AI agent that can chain multiple tool uses together and complete multi-step tasks independently.

Provided by

JetBrains Academy

About

AI agents can do more than just answer questions—they can actually help you with tasks. This project helps you build a simple but useful AI butler that works autonomously. Give it access to tools like checking the weather and managing a wardrobe, and watch it complete multi-step tasks on its own. No more single-question chatbots—see how AI can chain actions together to be genuinely helpful!

Graduate project

This project covers the core topics of the AI Agents: Theory and Practice course, making it sufficiently challenging to be a proud addition to your portfolio.

At least one graduate project is required to complete the course.

What you'll learn

Once you choose a project, we'll provide you with a study plan that includes all the necessary topics from your course to get it built. Here’s what awaits you:

Build a simple text-only chat with an LLM using a secure API setup and basic conversation flow.

Give your assistant real-world interaction by adding a weather tool and implementing the ACT phase of the THINK-ACT-OBSERVE loop.

Build a fully functional LLM agent that can use multiple tools with parameters, manage state, and maintain continuous conversation in a complete THINK-ACT-OBSERVE loop.

Reviews

John Yeany

2 months ago

5.0

This was a great project to get started with the think -> act -> observe cycle and taught me how to define tools for an LLM to call and how to handle the results.

Mateusz Duraj

3 months ago

5.0

I run into timeout error problem serveral times (in test is set to 15 seconds), I suggest to increase it, since sometimes open ai api has some issue with increasing latencies . Also instead of using open ai api, I tried to implement this simple agent with integration with local llm models (ollama + ...

Dustin Feucht

5 months ago

4.0

Often the "don't", "couldn't", etc. are not correctly recognized in the tests. Makes it really frustrating if the agent behaves correctly, but the tests fail.

4.5

Learners who completed this project within the AI Agents: Theory and Practice course rated it as follows:

Usefulness

4.7

Fun

4.5

Clarity

4.3