Inspiration
Teachers spend countless hours creating worksheets, organizing assignments, and grading student submissions. We wanted to build something that removes that workload completely — something fast, intuitive, and powerful enough to feel like a true assistant. AutoAssign was inspired by the idea that AI should save teachers time, not add complexity.
What it does
AutoAssign lets teachers:
- Organize assignments through a clean, IntelliJ-style planner
- Generate AI-powered worksheets with custom topics, difficulty levels, and question counts
- Export polished, classroom-ready PDFs
- Upload completed student worksheets and automatically generate AI-written performance summaries
- Seamlessly switch between planning, creating, and grading workflows
Everything happens inside a simple desktop app designed to feel fast, modern, and distraction-free.
How we built it
AutoAssign was built using:
- Java 17 for the core application
- Swing + FlatLaf for a modern UI experience
- Apache PDFBox for PDF creation
- Google Gemini API for worksheet generation and student performance analysis
- Gson for JSON parsing
- A hierarchical planner system modeled after IntelliJ’s project tree
We designed a full assignment workflow: planning → creation → editing → exporting → grading.
Challenges we ran into
- Ensuring the AI always returned clean, valid JSON (no LaTeX or markdown)
- Converting AI-generated text into a properly formatted PDF
- Designing a UI that feels modern despite using Swing
- Handling file exports, tree navigation, nested folders, and drag-and-drop
- Integrating the grading system without making the interface overwhelming
- Making loading transitions feel polished and responsive
Accomplishments that we're proud of
- A fully functioning AI worksheet generator with PDF output
- A robust planner system with nested folders and drag-and-drop organization
- Seamless navigation across the entire workflow
- A working AI grader capable of summarizing student performance
- A polished interface with smooth loading animations
- Building everything within a fast-paced hackathon timeframe
What we learned
- How to orchestrate multiple AI-driven components inside a desktop application
- How to structure a large Swing project while keeping it maintainable
- How to prompt AI models effectively for strict JSON output
- How to design intuitive UIs for educators (not developers!)
- The importance of polishing small details — like loading screens and naming — to make an app feel real
What's next for AutoAssign
- A complete Gradebook module with academic histories and class-level analytics
- Smarter worksheet templates and support for different subjects
- OCR-based student submission reading
- Cloud syncing for assignments across devices
- A web version for universal accessibility
- Teacher collaboration features and shared assignment libraries
AutoAssign started as a hackathon idea — now it has the potential to become a full classroom productivity platform.
Log in or sign up for Devpost to join the conversation.