📖 About the Project
What inspired me
During the AmpliCode Hackathon 2025, I wanted to create a tool that makes learning and note-taking smarter and faster. Reading through lengthy documents can be overwhelming, so I set out to build NoteGenius — an AI-powered app that transforms any document into concise summaries, key insights, and interactive quizzes.
What I learned
I learned a lot about integrating NLP models (like Mistral) into real-world applications, handling different file formats (.txt, .pdf, .docx), and creating a seamless user experience with FastAPI and Streamlit. I also discovered how to handle real-time AI responses and work with data parsing challenges.
How I built it
- The backend is powered by FastAPI and uses Ollama for AI-powered summarization and quiz generation.
- I implemented file parsing and text extraction with PyPDF2 and python-docx.
- The frontend is built with Streamlit for an intuitive user interface.
- I used session state to keep quizzes interactive and engaging!
Challenges I faced
- Handling large files and different file types (especially
.pdfvs.docx). - Making sure the AI model’s JSON responses were always parseable and accurate.
- Balancing concise summaries with meaningful insights from the model.
- Ensuring smooth communication between the frontend and backend, even during quick hackathon development cycles.
What’s next?
I’m excited to keep improving NoteGenius by adding support for more file types, refining the AI quiz generation, and polishing the UI for a better learning experience.
Log in or sign up for Devpost to join the conversation.