qCraft

The Adaptive Quiz App

qCraft is a text-based quiz generator that extracts content from PDFs, PPTX files, and scanned documents to generate quiz questions. It evaluates user responses and dynamically generates new quizzes based on areas of weakness, ensuring adaptive learning.

Features

• ✅ Extracts text from PDF, PPTX, and scanned documents (OCR)
• ✅ Generates multiple-choice and open-ended questions
• ✅ Evaluates user answers using AI-driven semantic matching
• ✅ Adapts quizzes based on user performance
• ✅ Works offline, with no reliance on OpenAI API

Project Structure

This project is divided into two repositories:

1. qEngine – Backend (Python + Machine Learning)
2. qCraft – Mobile App (Flutter)

Getting Started

Setting Up the Backend (qEngine)

1. Clone the repository:

   git clone https://github.com/Lukas-io/qEngine.git
   cd qEngine
   

2. Create a virtual environment:

   python -m venv venv
   
   source venv/bin/activate  # On Windows: venv\Scripts\activate
   

3. Install dependencies:

   pip install -r requirements.txt
   

Setting Up the Mobile App (qCraft)

1. Clone the repository:

   git clone https://github.com/Lukas-io/qCraft.git
   cd qCraft
   

2. Install Flutter dependencies:

   flutter pub get
   

3. Run the app:

   flutter run
   

Tech Stack

• Backend: Python, FastAPI, Hugging Face Transformers, Tesseract OCR
• Mobile: Flutter, Dart, GetIt (DI), BLoC (State Management)
• Database: Firebase/PostgreSQL (TBD)

Contributing

Contributions are welcome! Fork the repo and submit a PR.

License

This project is licensed under the MIT License.