AI Instructor – MCP

Teach live online classes on building MCP servers, so that AI apps and agents can reach tools and data through the Model Context Protocol, the open standard for those connections. You teach in English from a ready class kit, remotely, and you are paid for each class you teach.

What the work is for

YPAI Learn runs live classes where people learn to connect AI to the systems they already work with. Your classes take developers from a first local MCP server with one tool to a remote server behind OAuth sign-in that an agent can call, and teach them to check what each tool can read and change.

The work

• Teach the live class “Build an MCP server with tools and resources for your own data”, 90 minutes, from the class kit.
• Teach the live class “Secure a remote MCP server with OAuth and connect it to an agent”, 90 minutes, from the class kit.
• Show live how a tool's name, description and input schema shape when and how the model calls it, and how a resource that only supplies data differs from a tool that acts.
• Walk learners through the MCP Inspector: list a server's tools and resources, call one by hand, read the JSON-RPC messages and find out why a call failed.
• Help learners when a server will not start, a client cannot see it or a token is refused, and show where the server's own logs explain why.

What to expect

• Classes run in English, live online, for developers in many countries. The application also asks in which other languages you could teach the protocol.
• Each class kit holds a minute-by-minute plan, slides, speaker notes, a demo of an MCP server answering calls from an AI client, and a backup recording of that demo.
• You teach from your own computer, remotely, with a code editor and an MCP client set up for the demo, and a stable connection, a camera and a microphone.
• Before you teach an MCP class for the first time, you pass 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 MCP server you built or extended for real work, with code or a write-up you can walk through.
• Experience teaching developers, giving technical talks, writing tutorials or running hands-on workshops.
• Patience with developers who have called an API before but never written a server for an AI client.

You need

• Hands-on experience building or running MCP servers and connecting them to AI clients in your own work.
• Working knowledge of APIs, JSON and authentication, such as API keys and OAuth tokens.
• Fluent spoken English, the language the classes are taught in.
• 18 years or older.

Good to know

• If your employment contract asks for approval before you take outside work, get it before you accept an MCP teaching assignment.
• How often you teach MCP depends on demand for the classes; each invitation gives the expected time commitment.
• Applying costs nothing, and no YPAI Learn course has to be bought or taken first.