We started building Tulpy
We began developing Tulpy — a language-learning app built around a personalized feed of real content.
Every language app we tried eventually felt like homework. Flashcards drilled words we forgot the next day; grammar tables explained rules we could never quite use in a real sentence. So we started building Tulpy around a different idea: what if learning a language felt like scrolling a feed you actually wanted to read?
Why a feed, not a course
The research is surprisingly clear on this. Stephen Krashen's comprehensible input hypothesis argues that we acquire language when we understand messages slightly beyond our current level — what he calls "i+1". Not drills, not memorised rules: understandable, compelling content. Tulpy is built to deliver exactly that, one card at a time.
Instead of a fixed syllabus, you get a personalized stream of short stories, news, and facts tuned to your level and interests. You tap a word you don't know, long-press a sentence that's unclear, and like the pieces you enjoy. Every interaction teaches Tulpy a little more about where your "sweet spot" of difficulty actually is.
Remembering what you read
Comprehensible input gets you understanding fast, but understanding isn't the same as remembering. That's where spaced repetition comes in: Tulpy schedules the words and grammar patterns you've met to come back right before you'd forget them — the moment that strengthens memory most. You never build a deck by hand; the feed quietly does it for you.
Building in public
We're a small team and we'd rather show our work than polish a pitch. This blog is where we'll share what we're shipping, what we're learning about language acquisition, and the decisions (and mistakes) behind the product. If any of this resonates, follow along — the best feedback we get is from people learning alongside us.