KelimeYAP English - Learn Vocabulary

Learn English through movies and TV. Explore 20,000+ CEFR-aligned words, discover vocabulary from thousands of series and films, and master it with interactive quizzes and missions.

Comment 2

Inspiration

KelimeYAP started as a folded sheet of paper.

For years, I learned vocabulary the low-tech way: print words on an A4 sheet, fold it down the middle so the meanings stay hidden, and carry it everywhere. In the coffee line, on the bus, five minutes before a meeting, I would unfold a corner and try to remember the meaning before peeking. That small moment of recall is where the learning actually happens. I introduced the same system to many friends, and the results genuinely surprised all of us.

The designer in me wanted to turn it into a real product: physical flashcards with a scratch-off strip covering each meaning. I designed the cards and lined up manufacturers and suppliers. Then I ran into the fundamental flaw of physical scratch cards: you can scratch them exactly once. After that, the meaning stays permanently visible, and learning quietly turns into reading. Selling a fresh deck every time someone forgets a word might be a great business model, but it is a terrible learning experience.

Then it clicked. With nearly ten years of software development experience, I realized a phone could solve what paper and physical cards could not. The scratch-off layer could reset infinitely. Every card could arrive with its meaning hidden, every single time, without carrying folded papers or stacks of cards in your pocket.

The second idea came from how people actually consume entertainment. You don't need to stop watching what you love to learn a language. Explore vocabulary from a film you plan to watch or an episode of your favorite series during breakfast, on your commute, or during a lunch break. By the time you press play, you have already encountered the words.

That combination became KelimeYAP EN: turning everyday vocabulary learning into something engaging, repeatable, and connected to the content people already love.

What it does

KelimeYAP EN is a swipe-based English vocabulary learning app designed primarily for Turkish speakers, combining active recall, personalized vocabulary tracking, and entertainment-based learning.

  • Swipe to decide. Swipe right for "I know it", left for "I don't", and down for "almost there". Every swipe contributes to a confidence score, making progress measurable rather than guessed. Meanings stay hidden beneath an interactive scratch-off layer that resets whenever a card returns.
  • 20,000+ vocabulary words. A comprehensive collection spanning CEFR levels A1 to C2, drawing from Oxford 3000/5000, Longman 3000 and OPAL, with example sentences and 5,094 illustrated hint scenes. More than 540 collocations and useful phrases provide additional context.
  • Learn through movies and TV shows. Explore vocabulary decks from tens of thousands of films and thousands of TV series, including their seasons and episodes. Every deck shows how many words are new to you and automatically excludes words you already know. Prepare for tonight's episode or an upcoming film by learning its vocabulary beforehand. The app uses vocabulary only, never copyrighted dialogue, subtitles or film transcripts.
  • Knowledge must be demonstrated. Words marked as known enter a mission pool. Each mission consists of six stages and a final challenge, featuring 80 questions across seven question types, including picture-based questions. A word leaves the pool only after being answered without a mistake. Incorrect answers return as personalized revision quizzes.
  • Progress without pressure. A mountain journey represents the learning path from A1 to C2, accompanied by badges, puzzles and achievements. Notifications are limited to a maximum of two per day, and every notification category can be disabled.
  • Kulabuz, your in-app guide. A friendly companion that answers questions about the application, powered by Claude through a Supabase Edge Function. No API keys or prompts are stored on the device. It is the app's only live AI integration: learning content is generated beforehand, and the core learning experience works offline.
  • No sign-up, ever. An anonymous identity is created on first launch. Learning progress can be transferred to another device using a transfer code, and users can permanently delete their server-side learning data through the in-app deletion option.

How we built it

KelimeYAP EN is developed by Obliquex Global, a US-based, Turkish-founded software team of six.

The mobile application is built with React Native and Expo using the managed workflow, Expo Router and TypeScript.

The backend runs on Supabase, using PostgreSQL, Anonymous Auth, Storage and Deno-based Edge Functions for operations that should not be trusted to the client.

On-device vocabulary data is managed through expo-sqlite, with batched delta synchronization in both directions. Every swipe is saved locally first and synchronized with the server in periodic batches. The card interactions, scratch reveal and mountain map are powered by react-native-reanimated and react-native-gesture-handler.

The 25 sound effects are procedurally synthesized through a custom script rather than sourced from a conventional sound library.

The app supports Turkish and English, while EAS Update allows JavaScript updates over the air with fingerprint-based runtime matching.

Monetization has two independent layers.

RevenueCat powers KelimeYAP Pro, with monthly, quarterly and yearly subscriptions across both app stores. The paywall can be remotely configured without releasing a new app version, and includes a 50% first-month introductory offer.

AdMob supports free users through native advertising (one ad per 20 cards), an optional rewarded video and a single banner inside the example-sentence sheet.

There are no interstitial advertisements. Rewarded credits are granted only after an AdMob Server-Side Verification signature has been validated by a Supabase Edge Function using Google's ECDSA P-256 public keys.

Learning content, including definitions, example sentences, image prompts, quiz captions and entertainment-based deck word lists, is generated in bulk beforehand using Claude Code and stored directly in the database. No part of the core learning loop depends on a live AI request.

Challenges we ran into

  • Making the credit economy fair. Our initial model charged credits for opening example and bonus panels. Testers were charged again whenever they reopened a panel on the same card. We redesigned the system around one credit per card, making every panel free to reopen.
  • Establishing proper trust boundaries. The first rewarded-video endpoint trusted the client to provide the credit amount. We rebuilt it around AdMob Server-Side Verification. A Supabase Edge Function now validates Google's ECDSA P-256 signature against published keys before granting any credits.
  • A bug that looked like a feature. The mission mistake list was always one step behind because it read React state in the same tick it was updated. The list was never empty, just incorrect, allowing the issue to survive initial device testing until we traced the state-update behavior.
  • App Store configuration drift. Our three subscriptions were accidentally available in 64, 63 and 38 countries respectively. We aligned all three with the application's worldwide availability.
  • Anonymous identity across platforms. On iOS, the session lives in the Keychain and survives a reinstall; on Android, it does not. We verified this with real tester data and designed the transfer-code mechanism to handle the platform differences.

Accomplishments that we're proud of

  • A vocabulary ecosystem with 20,000+ words, 5,094 illustrated scenes, and a comprehensive collection of movie and TV vocabulary decks.
  • An 80-question mission system spanning seven question types, designed to test actual knowledge rather than simple familiarity.
  • A core learning experience that is never blocked by an advertisement or paywall. Pro removes ads and adds convenience; it never locks the ability to learn.
  • A genuinely frictionless start: no account creation, email verification or login screen before the first card.
  • An offline-capable learning experience with local-first persistence and background synchronization.
  • Turning a simple folded A4 sheet of paper into a complete mobile learning ecosystem.

What we learned

  • App Store and Google Play sandboxes can behave differently from real purchasing environments. For example, TestFlight displayed USD in the paywall while the actual purchase sheet showed local currency. We learned to verify critical experiences on their real production surfaces.
  • Fair monetization requires as much iteration as the learning experience itself. A credit system that looked reasonable on paper behaved very differently when real users interacted with it.
  • Anonymous-first identity dramatically reduces onboarding friction, but introduces engineering challenges around device migration, reinstalls, session persistence and data deletion.
  • Generating learning content in advance creates a more reliable and predictable experience. It reduces runtime dependencies, avoids unnecessary AI costs and allows the core learning experience to remain available offline.

What's next for KelimeYAP EN

We are continuing to refine the credit economy and advertising frequency using remotely configurable settings, without requiring new app releases.

Our roadmap includes privacy-respecting product analytics, additional vocabulary deck categories beyond films and television, and code-based live vocabulary duels between friends.

We also want to keep expanding the connection between language learning and everyday entertainment, making it easier for learners to turn the content they already enjoy into opportunities to improve their English.

Built by Obliquex Global.

Built With

Share this project:

Updates

posted an update —

Hey guys, I'm not sure if I properly "Submitted" the project. Reach me out anytime, I put serious effort to this one and genuinely want to compete in the shipaton:) Sorry for the ambigiouty, I'm lacking sleep and can't remember if it looks draft or a proper submission