Inspiration
We wanted to move away from the flat, sterile design of modern software and bring back the joyful, optimistic aesthetic of Frutiger Aero UI - glass panels and a calm UI.
What it does
Our tool is simple and intuitive to use.
PaperTrace helps you review complex documents (PDFs or DOCX files) without losing your place. It extracts requirements into an interactive checklist side-by-side with a live local document viewer. As you work through requirements, it tracks your progress, shows original source text on demand, links conditional follow-ups, and exports a report with an action list.
How we built it
We had to use generative AI to guide us through the process of creating the website.
Backend: Built with Python using FastAPI and local processing libraries to parse documents and structure requirements alongside with local AI llama3.2
Frontend: Built with JavaScript and React, featuring custom glassmorphism styles, fluid layout resizing, memoized components for high performance, and Lucide icons.
What's next for PaperTrace
In future, we might try to create an even faster and smarter local AI model, making document parsing and requirement mapping completely offline and instantaneous.
Please be aware that the backend is a mock up of what is possible and that we have tried our best in creating an original UI with a comfortable UX.
Built With
- fastapi
- javascript
- lucide
- ollama
- python
- react
Log in or sign up for Devpost to join the conversation.