My buddy's in med school. He spends more time formatting study materials than actually studying.
Last week he sent me a 30-page lecture PDF. Said he'd been up until 2am making flashcards the night before his last exam.
I pasted it into NoteFlux. The translation model ran first (his lecture was in French), the segmentation model split it into topics, then the generation model created Q&A pairs for each one.
5 minutes later he had 47 flashcards ready to review.
His response: "I want to punch myself in the face for doing that manually for two years."
That's why I built it. Not to replace studying — to replace the busywork around studying.
no API keys for users to manage. Paste text or upload audio → flashcards appear. That's the whole pitch.
One thing I find interesting about tools like this is that they don't just remove effort.
They also remove a step that used to force interaction with the material.
Sometimes that's obviously a win.
Sometimes the busywork was quietly doing a second job the user never intended it to do.
That tension ended up feeling more interesting to me than the time savings themselves.
There's a feature I've implemented in my NoteFlux app. It's called "feed page" at the moment and it's main purpose is to let you answer through your generated cards. It will let you know whether you've answered correctly or wrong, if you've answered incorrectly you will not be able to go forward, the app lets you skip the question if you're really stuck on it but other than that you're restricted to just answering the current card
That's actually the part I'd be most curious about.
Whether the restriction is improving memory, or whether it's changing behavior in a way that indirectly improves memory.
What's the best email to reach you on?
The restriction isn't perfect, but it works when used properly. I'm definitely looking forward to improve it though.
You can reach me on zairysbigtae@gmail. com
Appreciate it.
Just sent you an email.
The distinction ended up feeling more important to me than the feature itself.