AI Instructor – AI agents

Teach live online classes for developers on building AI agents in code: tool definitions, the agent loop, planning, memory, evaluation and guardrails. You teach in English, remotely, with a class kit prepared in advance, and you are paid for each class you teach.

What the work is for

YPAI Learn teaches developers in live classes to build AI agents in code. Your classes take them from a model that calls one function to an agent that plans a multi-step task, keeps state and uses several tools, and teach them to check its trajectories and stop unsafe actions before anything runs for real.

The work

• Teach the live class “Code an agent that plans, calls tools and remembers across steps”, 90 minutes, from the class kit.
• Teach the live class “Evaluate agent trajectories and add guardrails to risky tool calls”, 90 minutes, from the class kit.
• Show live how tool names, descriptions and argument schemas steer which tool the model picks, and how the loop feeds each result back until the task is done or a step limit is hit.
• Walk learners through tracing a run step by step, writing test tasks with expected outcomes, and scoring both the final answer and the path the agent took to reach it.
• Help learners when an agent loops, calls the wrong tool or invents arguments, by reading the trace and tightening tool descriptions, permissions and approval steps.

What to expect

• Sessions happen live on video, in English, for an international group of learners. The application also lets you name other languages you teach in.
• In the kit for each class: a minute-by-minute plan, slides, speaker notes, a demo agent with its traces, and a recorded backup of the demo.
• From your own computer, you teach remotely with a camera, a microphone and a stable connection, and a development environment where the demo agent runs.
• You teach a class once you have passed the instructor test for that class.
• Before a teaching assignment starts, you sign an instructor agreement. It covers pay, confidentiality and how class material may be used.
• Each invitation states the class times and the price learners pay, and you see both before you accept.

What you bring

• An agent you built in code for real work, with its repository, traces or evaluation results that you can explain.
• Experience teaching programming, mentoring engineers, running code labs or reviewing other people's system designs.
• Patience with developers who expect an agent to behave like a fixed script.

You need

• Hands-on work building AI agents in code: tool calling, state and memory, multi-step planning, tracing and tests.
• Strong programming in Python or TypeScript, including async code, error handling and writing tests.
• Fluent spoken English, the language the classes are taught in.
• 18 years or older.

Good to know

• An employment contract may require approval for outside work. If yours does, settle it before you accept an assignment.
• How many classes come your way depends on demand; the invitation for each states the time it expects you to commit.
• The application is free, and you do not need to have bought or taken a YPAI Learn course.