All tutors
Build your first AI agent
How agents use tools in a loop, and how to build one that stays safe.
What you will be able to do
- Explain the agent loop: model, tools, observations, stop condition
- Define a tool with a name, a description and an input schema
- Build a small agent that calls one or two tools
- Add limits, logging and human approval for risky actions
- Evaluate an agent with a set of test tasks
Lesson plan
- 1 What makes something an agent Tell an agent apart from a single prompt or a fixed workflow. Start
- 2 Tool calling Understand how a model asks your code to run a tool. Start
- 3 The loop in code Write the basic agent loop and its stop conditions. Start
- 4 Connecting tools with MCP Know what the Model Context Protocol offers. Start
- 5 Safety and limits Keep an agent from doing damage. Start
- 6 Testing and improving an agent Measure whether the agent works before trusting it. Start
Try asking
About this tutor
For people who know basic Python or are comfortable reading code. You will learn what makes an AI system an agent, how tool calling works, and how to design, test and limit a small agent that does one job well.
Reviews
Students can review a tutor after a paid lesson. Nobody has yet.
About the teacher
Teaches machine learning, Python and AI agents
3 tutors
I teach the technical side of AI to people who are new to code or to machine learning. My tutors explain the idea first, then show a small working example, then ask you to change it and see what happens.
See Daniel's profile and tutorsMore like this
Other tutors on the same or nearby topics.
The Tool Calling Loop, Explained Step by Step