You build something with it.
You and the tutor build something real together in your own project, such as a prompt pack, a checklist or a skill of your own. Then you keep it.
AI Foundations is coming. Register your interest.
Tutr turns the AI agent you already use into a one-to-one tutor.
Why one to one
In 1984 Benjamin Bloom compared students taught in a class with students taught by a tutor. The average tutored student did better than 98% of the class.
The method was never the problem. The cost was. Your AI agent already sits beside your work, so Tutr gives it a course to teach and the habits of a good tutor.
The same way of teaching in every Tutr course.
You and the tutor build something real together in your own project, such as a prompt pack, a checklist or a skill of your own. Then you keep it.
When a choice is yours, the tutor asks, then waits for your answer. A decision the tutor makes never counts as yours.
A lesson counts as passed only when your saved work passes its check, not when the chat feels finished.
Thirteen lessons for anyone who uses AI in their work, whether or not you write code. The course is coming, and it isn’t available yet.
Coming to Tutr
You’ll learn how AI agents work by building with one: write requests it gets right, decide what it may see and do, connect an app, and teach it a skill of your own.
Getting started3 lessons
What a model is, what your AI can see and do, and the approvals you set first.
Asking well3 lessons
Requests it gets right first time, and a prompt pack for jobs you repeat.
Context and trust3 lessons
What to share, what to keep private, and how to catch a confident mistake.
Connect an app2 lessons
Give your AI one of your apps, read-only first.
Your skill, your call2 lessons
Teach it a job you repeat, test it, and explain what you kept for yourself.
Outline subject to change.
Tutr is free and open source, and you never need an account to learn. You bring the AI agent you already use.
Tutr never receives your files or your conversations. They stay where you run your agent, and nothing goes anywhere else unless you opt in.
Your progress shows only what passed its check. Skipped work never shows as passed, and the tutor’s decisions never count as yours.
Tutr is open source under Apache 2.0, and its courses use the open Agent Courses format. You bring the agent and the model.