⚡ Unflash

🌎 Background

We can all agree that flashcards are one of the best ways to study. However we realized that they have an overlooked flaw.

Flashcards train your brain to look for specific words in definitions to tie to words. This means that they only teach you to remember, not understand. Unflash uses Artificial Intelligence to reword definitions without changing their overall meaning.

Ultimately, the purpose of Unflash is to help people tie meanings to words, rather than words to words.

❓What is Unflash?

Unflash presents the concept of Unflashcards: flashcards that help you memorize meanings, not definitions.

Big Brain

After seeing a definition and correctly recognizing it, Unflash will utilize the OpenAI GPT-3 model to reword the definition you are presented with. This helps build on your understanding of the topics as you have to interpret the meaning of the words and not the words themselves.

🚀 Technologies used

Unflash is built with Flutter, a framework for developing mobile apps that work on Android and iOS devices. Firebase Auth and Firestore are used to handle authentication and storing user data and sets.

💥 Challenges we ran into

Our app is highly asynchronous, whether it is database queries, or API requests, asynchronous code came with its own implications.

✨Accomplishments

We put a lot of effort into our UI, and it turned out great! While the actual app fell a bit behind the prototype, we're proud of where we got regardless.

More notably, this is our first project that utilizes artificial intelligence. Fortunately, we got it working without many setbacks.

🎓What we learned

We learned how we can leverage the power of the GPT-3 AI model to fuel projects that otherwise would have been a dream.

We've also gained valuable knowledge in prototyping (hi-fi and lo-fi), dealing with Flutter, and dealing with Firebase spontaneously breaking for no reason.

🌇 What's next

We hope to fine tune our model to be more concise and to perform better at rewording sentences. Since we are students, we can easily find ourselves using this in our daily lives, further motivating development. Perhaps one day this app will become mature enough to become available on App Stores, something we have been dreaming of for a long time.