Update your
mental weights.
Passively reading a solution feels like learning and mostly isn’t. Retention comes from recall under friction — pulling a derivation out of your own head, getting it wrong, seeing exactly where it broke, and writing it again.
Session 6 of 24
A stock's daily return has mean 0.1% and variance 0.0004. What is Var of the 5-day return, i.i.d.?
1 Var(R₅) = Var(R₁ + R₂ + … + R₅)
2 Cov(Rᵢ, Rⱼ) = 0 for i ≠ j (i.i.d.)
3 Var(R₅) = 0.0004
Mastery, decaying
Session 3 of 20
A jar has 12 red and 8 blue marbles. Two are drawn without replacement. What is P(both red)?
1 P(R₁) = 12/20
2 P(R₂ | R₁) = 11/19
3 P(both) = 12/20 × 12/20 = 9/25
Mastery, decaying
Session 5 of 31
Given an array of integers, return indices of the two numbers that add up to a target.
vector<int> twoSum(vector<int>& nums, int target) {
unordered_map<int, int> seen;
for (int i = 0; i < nums.size(); i++) {
if (seen.count(nums[i])) return {seen[nums[i]], i};
seen[nums[i]] = i;
}
return {};
}
Mastery, decaying
How it works
A method, not a feature list.
You read, then you write from memory.
Each session has a pre-read — a handout derived from first principles, questions and answers included. Read it beforehand; ask it anything in the margin. Then the notebook opens, the handout is gone, and a question appears with nothing but blank page underneath.
Errors are located, not solved.
When a derivation goes wrong, the tutor highlights the exact line where it broke and pins a short note to it — the misconception, not the fix. You still have to find the flaw.
Getting it wrong is the mechanism.
Re-derive under the correction. The physical act of rewriting is what moves it into memory, so the page is built to be written on repeatedly rather than filled in once.
It plans the shape first, content later.
You say what you're preparing for and how many hours you actually have; the syllabus is fitted to that budget up front — and what got dropped to make it fit is written down, not silently cut.
Four doorways
One thing you can prepare for.
The first three are one machine with different intake. Only reading is different — the one you can use in thirty seconds with no setup at all.
Interview prep
A loop at a named company by a known date. Paste the job description and the syllabus is built from what the role actually tests.
Exams
The GMAT, as GMAC publishes it — every section, in the official order. More exams are being added one at a time.
Custom prep
A topic, a link, or a syllabus you paste. Worked through one sitting at a time, derived rather than recited.
Reading
Any PDF, or a blank page. No syllabus, no score — somewhere to read and ask.
Most of it happens right here
Plan a syllabus, work a session as a live conversation, write code with real syntax highlighting, check progress, read back a past session — all in the browser. No app required for any of it.
iPad supported, for handwritten work
The notebook — pencil, page, the tutor watching your ink — is a native iPad app. A derivation wants a pen sometimes, and nothing about typing replicates that; iPad’s there when it does.