loading…
Quiz 1 · MLF & MLT & Maths 1 & Mathematics for Data Science II & Statistics for Data Science II & Statistics for Data Science 1 · in --d --h
Interactive lessons rebuilt from your real lectures. A tutor that cites the professor to the second. Real past-year questions, graded instantly. Built by a student sitting the same exams.
What it does
Every concept is rebuilt from the actual lecture transcript into an interactive lesson: you predict, you see the picture, then the notation arrives, in that order. No wall of definitions, no rewatching a 40-minute video for one idea.
Built from the real transcripts, never a summary of a summary.
Predict before any math
500 data points, two dimensions each. You're only allowed to keep one number per point. Squash them onto which line to lose the least?
Ask any doubt. The tutor answers only from the actual lectures, and every claim carries a chip with the lecture and the second it came from. Tap a chip, and the official video opens at that exact moment. If it can't find a source, it says so.
Try the chips below. They really do open the lecture at that second.
Real past-year questions are tagged to every concept, and an AI generator produces fresh variants in the same style: MCQ, MSQ, NAT, graded instantly, with wrong answers feeding back into what you revise next.
The demo below runs on real Week-4 estimation questions.
In Bayesian estimation, consider a Beta prior for the parameter p of a Bernoulli distribution: p ~ Beta(4, 3). The dataset has 8 ones and 5 zeros. What is the posterior?
Why you can trust it
Everything on this site is built from one source of truth, the official lectures, through a pipeline that never lets a claim detach from where it was taught.
Official lectures
The professor's actual videos, week by week.
Timestamped transcript
Every paragraph pinned to its second in the video.
Concept lessons
Rebuilt into predict-first interactive lessons.
Cited answers & PYQs
Tutor replies and questions that point back to the source.
Fair questions
“The lectures are already free on YouTube.”
They are, and they're the source of everything here. The difference is 40 hours of video versus the one moment you need, found in seconds, with the link to jump straight there.
“I can just ask ChatGPT.”
It hasn't watched your lectures. It will answer confidently in the professor's voice about things the professor never said, and you won't know which parts. Here, every sentence is grounded in the transcript or it isn't said.
“My seniors' notes are enough.”
Notes tell you what someone else understood. They can't quiz you, can't grade your own-words explanation, and can't tell you which Week-3 idea you specifically keep fumbling.
What's next
Two courses are live. The roadmap isn't a suggestion box, the most-voted course genuinely gets built next.
ML Foundations and ML Techniques are live. Vote for the next course, the winner genuinely gets built next.
First, where are you in the program?
Right now, every check-yourself score is instant and per-lesson. Next: a mastery map that carries across weeks, and a tutor that remembers where you've struggled, so a doubt about “conditional probability” in week 6 gets an answer that references the gap it saw in week 3.
Questions
One more thing
We just made them searchable, teachable, and testable.
Start Quiz 1 prep →Free to start · no card · built by a BS student sitting the same Quiz 1