How Norudit Actually Personalises Your Learning
"Personalised" gets used loosely. For most apps it means one thing: a difficulty slider that nudges up or down based on your last few answers. Norudit means something closer to the literal definition, it's actually building a model of you, across several different dimensions, and using all of it.
What you're interested in
The first is interest. In settings, you can list up to 15 things you're actually into, and Phase 2 of the Norudit Sequence maps whatever concept you're learning onto one of them, but only where it genuinely fits. Nothing gets forced: if none of your interests map cleanly onto the concept, it falls back to a general analogy most people would relate to instead. Where it does fit, force in physics can become "how much force would Superman need to lift the Earth" if you're into comics, or a tax lesson in economics can become "how would you tax a kingdom you just got transmigrated into" if you're into manhwa. Same concept, wrapped in something you'd actually want to think about. I've written about the mechanics of this, and why it works, in Why Grades Shouldn't Be the End Goal of Studying.
The profile it builds of you over time
Interest is just the input. What happens after is a profile that keeps building the more you use the app. Every session leaves a trace, what you struggled with, what clicked instantly, when you tend to actually focus best, and those don't just sit as raw logs. They get consolidated into facts your account can actually use later: "bond enthalpy calculations keep slipping," "you passed the equilibrium gate on the second try," "your strongest sessions are mornings." That profile is what lets the app surface the right thing at the right time, an exam reminder that actually says what you're weak on, instead of a generic "don't forget to revise."
This is also grounded, the same discipline as the rest of the app: the profile is built from your own material and your own answers, not a guess about "students like you." More on how Norudit stays grounded to your material specifically here.
Your forgetting curve, not a textbook average
Spacing is personalised too. Norudit schedules reviews with FSRS-6, but the version that actually matters is that the algorithm refits its own parameters to your review history over time, so the spacing schedule ends up modelling how you specifically forget, not a hypothetical average student. Two people studying the same class end up on genuinely different review timetables, because their actual recall patterns are different. The full story on how the memory layer fits itself to you is here.
Your mastery, concept by concept
Underneath all of it is Bayesian Knowledge Tracing, keeping a live, per-concept estimate of what you've actually mastered, not just what class you're in or what topic you're "on." That estimate drives more than just what gets reviewed next:
- Concept-varied repetition varies a review card's difficulty based on how well you've mastered that specific concept, not the same fixed card everyone else gets.
- Difficulty steering adapts to both your last grade and your mastery: struggle or a lapse pulls a card back to consolidation, strong mastery pushes it toward harder, applied, edge-case questions instead of just repeating the same thing.
Nobody else studying the same textbook is getting the same sequence of cards. I go into the BKT mechanics and the spacing science together in Built on the Science of Memory, and the card side specifically in The Deck That Rewrites Itself.
The smaller personalisation
Not everything here is about learning mechanics. Lura, the companion that runs alongside your studying, has a persona system too, an assistant that looks like you and is voiced like you, not one default face for every user. It's a smaller thing than the mastery model or the memory layer, but it's part of the same instinct: nothing about this app should feel built for a generic student, because there isn't one.
Read more
- The five principles the whole method is built on: Built on the Science of Memory
- Why relatability is the mechanism behind interest-based mapping: Why Grades Shouldn't Be the End Goal of Studying
- How FSRS refits itself to your personal forgetting curve: The Academy That Remembers You
- How the cards themselves adapt to your concept mastery: The Deck That Rewrites Itself
- How the AI stays grounded to your own material: Grounded on the Page