Brainback
researchproduct design

The anti-flashcard

Why we generate recall cards from your errors, not from pre-made decks.

Aug 11, 20264 min readBrainback Research
abstract

Pre-made flashcard decks optimize for someone else's gaps. Brainback recall cards optimize for yours.

The flashcard is the oldest study tool that survives to the present. It works. It also fails, quietly and often, when the deck is not yours.

A pre-made deck optimizes for the average learner's gaps. Yours are not average. They are yours. A card that tests knowledge you already have is a card that costs you time and gives you nothing. A card that skips knowledge you don't have is a card that reinforces a gap.

Brainback recall cards are generated from problems you got wrong, or from concepts you explained back poorly. The prompt is rephrased so you can't rely on surface recognition, but the concept is one you failed to demonstrate. Every review is a test of a gap you actually have.

This is more expensive than a shared deck. It requires a Haiku call per card, versus a static database lookup. That is a tradeoff we make gladly. Your gaps are worth more than the delta in inference cost.


Written by Brainback Research. Published Aug 11, 2026. Filed under Product design.

if this landed, share it with the friend who’s outsourcing their brain

if the research landed

Put it into practice this semester.

Every essay here is downstream of a design decision inside Brainback. Come see the design.