SteadyStudy | Devpost

Launch Screen (Click on photos to view in highest quality)
Login Screen
Study Assistant Results
Agenda View
User Preferences Page
Agenda View (Light Mode)

Inspiration

The inspiration for this project was the pressure of academic buildup over the course of the semester, and how procrastination plays a role in creating stressful situations. The goal of this project is to aid a wide variety students with time management, from those who have mental disorders to those who simply don't have experience with the large course loads college can bring.

What it does

Our app solves student procrastination by transforming overwhelming assignments into manageable, automatically-scheduled work sessions. Instead of staring at "write a 10-page research paper," students get structured plans directly added to their Google Calendar.

How we built it

We built SteadyStudy as a full-stack application with a clear separation between frontend and backend. We built the frontend using Next.js 15 with the App Router and React Server Components, React Big Calendar with drag-and-drop, Framer Motion animations, and Tailwind-based dark/light theming. The backend is a FastAPI service with WebSocket endpoints and a custom function-execution framework that lets the Gemini 2.0 Flash Thinking model call seven actions (create assignments, schedule tasks, update calendar, etc.), plus PDF parsing with pypdf and MongoDB via Motor for flexible data modeling. We also added an integration layer that handles Google OAuth 2.0, JWT auth with bcrypt, and a scheduling algorithm that respects user availability, study times, and subject difficulty, using the last 20 chat messages to keep the AI context-aware without hitting token limits.