An Anki alternative that writes the cards.
If you tried Anki and stopped, it almost certainly was not discipline. It was the evening you had to spend turning one lecture into forty cards, every time, before the method gave you anything back.
Start with what Anki gets right
Anki is a serious piece of software and it is free. Two decades of scheduling work, an add-on ecosystem for almost anything, shared decks numbering in the millions, and total control over how a card behaves. If you want to configure the last knob of your review queue, nothing else comes close.
The method is not in question either. Retrieval beats rereading, and spacing the retrievals beats doing them together — that issettled research, not a product claim. Anki implements it faithfully.
The bill nobody mentions
The catch is that Anki assumes the cards already exist. Making them is a real skill: splitting an idea into single facts, phrasing each as a question with exactly one answer, keeping the wording stable enough that you are testing your memory and not your reading. Do it badly and you drill ambiguous questions for months.
You can see how that lands in the population most committed to Anki. In a 2025 survey of first-year medical students at the University of Central Florida, 97.6% of Anki users studied from pre-made decks rather than cards they had written. Out of 84 users, two made their own. These are people who use Anki daily for years and whose exams depend on it — and even they route around the card-writing.
That is the honest shape of the problem. The method is excellent and the entry fee is an evening you do not have.
What changes here
badNerd starts one step earlier: not with study material you have already decided to study, but with the thing you were watching anyway. A conference talk, an article someone sent you, a video you opened at midnight. You save the link; the AI reads it, pulls out what is worth keeping, writes the questions, and schedules them.
That distinction matters more than it sounds. Every other tool in this category asks you to upload a PDF or a lecture — which means you have already made the decision to sit down and study. Most of what you learn in a working year never gets that decision. It is the talk you meant to revisit, and it is gone by Thursday.
Each card keeps a link to the exact place it came from — the timecode in the video, the paragraph in the article — so verifying one takes a second, and you can edit any of them. Around it sits the rest of the source: the summary, the full transcript, a chat that only knows this video, and your own notes. Not five apps. One object.
What stays the same
- Spaced repetition with expanding intervals, and failed cards going back to the start.
- Retrieval, not recognition — reviews ask you to produce the answer, including typing it out for vocabulary, with the missed letter marked.
- A daily queue that is short by design rather than by luck.
- Your data as yours: export everything to Markdown, any time.
What you give up
Worth being straight about, because the answer is not nothing.
- No add-on ecosystem. Anki's is enormous and some people depend on it.
- No shared community decks. If your field already has a curated deck that thousands of people have refined, that head start is real and badNerd does not have it.
- No offline desktop app. badNerd runs in the browser and installs like a native app, which is not the same thing.
- Less control. No custom card types, no image occlusion, no scheduler you can tune by hand.
Who should stay on Anki
If you are a medical student with a deck like AnKing already curated for your exams, stay. The head start beats anything generated. If you need offline, or add-ons, or you enjoy building cards — that enjoyment is genuine and it solves the whole problem — stay.
badNerd is for the other case: you are not studying for a date, you are trying to keep what you already spend hours consuming, and the reason it is not working is that nobody sustains an evening of card-writing for a talk they watched on a Tuesday.
Get started freeFree while in beta · nothing to install
Sources
- Utilization Patterns and Perceptions of Anki Among First-Year Medical Students — 89 surveyed, 84 Anki users, 97.6% on pre-made decks. University of Central Florida College of Medicine, published 2025.
- Roediger & Karpicke (2006) — repeated retrieval beats repeated study at one week.
- Cepeda et al. (2006) — meta-analysis of distributed practice.