A tutor that argues with itself, on purpose.
Three AI tutors that disagree productively, show the reasoning, and help students see why one answer is better than another.
Because the inner loop you would naively write is replaced by a binary search on the patience-sorting structure. Each insertion is log n, and you do n of them.
Patience sorting analogy is the cleanest mental model. Imagine dealing cards into the smallest stack with a top card greater than yours. The number of stacks is your answer, and the binary search is the speed-up.
If you trace it on the example [10, 9, 2, 5, 3, 7, 101, 18], you see only 4 stacks form. That visualisation is what makes the n log n click for most students.
Patience sorting + binary search. Walk through the example [10, 9, 2, 5, 3, 7, 101, 18] step by step; you will land on 4 stacks. Each insertion is log n, total is n log n. The visualisation is the unlock.
What changes for you
Students see why one answer is better, not just what it is. The reasoning is the lesson.
Three angles on the same concept. One sticks for one student; another sticks for another.
Verdicts pull back to sources so the answer can be checked, not just trusted.
Which mind to ask
Each model leans differently. Maya keeps the differences visible.
Best for clear long-form explanations, careful reasoning chains, and humanities prose.
Best for STEM problem-solving, code, and structured walk-throughs.
Best for visual explanations, image-based problems, and quick fact-checks.
Where teams reach for it
Hear from three minds, get one verdict.
New accounts get $1 of trial credit. No card required to begin.