About
Built in London, Ontario by people who mark math.
Why we built Markscanner
Grading a class set of handwritten math takes too long, and the part that takes longest is the part nobody wants: the repetitive first pass where you’re mostly writing the same comments on the same mistakes for the twentieth time. We built Markscanner to take that first pass off the teacher’s plate, so the time left goes to the judgement calls and the conversations that only a teacher can have.
What we believe about AI in grading
AI should draft, not decide. The model suggests, the teacher reviews and approves, and nothing gets returned to a student until a real person has signed off on it.
Markscanner is deliberately designed as a teacher-in-the-loop tool. Low-confidence answers get flagged; they don’t get silently guessed.
We don’t use student work to train models. The content goes to the grader, does its job, and that’s where it stops.
Who we are
Markscanner is built by a small team that includes practising math teachers. We’re based in London, Ontario, and the tool is shaped around the way math is actually taught and assessed in Canadian classrooms — partial credit, worked steps, rubric marking, and paper-based quizzes and chapter tests.
If you’ve got feedback, a feature request, or a weird edge case from your own marking, we want to hear it. The product gets better when teachers push on it.