Your AI companion that breaks down research papers so you can understand them faster and with clarity
PaperBuddy is a Streamlit web application that helps users read and understand research papers using an OpenAI model. It uses a step-by-step teaching approach to guide learners through complex academic concepts.
- 📄 PDF Upload: Upload any research paper in PDF format
- 💬 Interactive Chat: Ask questions and have a conversation about the paper
- 🎓 Step-by-Step Learning: The AI tutor guides you through concepts incrementally
- 🔄 Clear Chat: Reset the conversation to start fresh
- 🤖 Powered by OpenAI: Uses GPT-5-mini for intelligent responses
- Clone the repository:
git clone https://github.com/nerdjerry/PaperBuddy.git
cd PaperBuddy- Install dependencies:
pip install -r requirements.txt- Set up your OpenAI API key:
export OPENAI_API_KEY='your-api-key-here'- Start the Streamlit app:
streamlit run app.py-
Open your browser to the URL shown (typically http://localhost:8501)
-
Upload a PDF research paper
-
Start asking questions!
- PDF Processing: When you upload a PDF, PaperBuddy converts it to clean markdown using
pymupdf4llm - AI Tutor Setup: The paper content is fed to an OpenAI model with a specialized system prompt
- Interactive Learning: The AI follows a teaching methodology:
- Assesses your background knowledge
- Builds intuition with examples
- Connects concepts to math
- Lets you guide the conversation
The AI tutor follows these guidelines:
- Takes one small step at a time
- Asks what you already know before explaining
- Keeps responses short (2-4 paragraphs max)
- Uses concrete examples and analogies
- Teaches math through code experiments when possible
- Always ends with a question to check understanding
- Python 3.8+
- OpenAI API key
- Dependencies listed in
requirements.txt
Feel free to open issues or submit pull requests!
MIT