CTRL-HACK-DEL 2.0 Project

Study Sphere is an 'AI-powered study partner matching platform' designed to connect students who are academically compatible. Instead of randomly forming study groups or hoping to bump into someone in your major, Study Sphere uses intelligent matching algorithms to find your best study partners.

The platform helps students:

Find study partners within their university and major Get AI-powered matches based on academic compatibility Connect with others in the same faculty, major, and year Stay accountable and motivated through collaboration Strengthen understanding by studying with the right people Build meaningful friendships while improving academic performance How we built it We built Study Sphere as a full-stack web application using the 'MERN stack' (MongoDB, Express.js, React, Node.js) with 'Google Gemini AI' integration.

Frontend (React) Built with React Implemented React Router for seamless navigation between pages Used React hooks (useState, useEffect) for state management Created custom CSS with modern, clean design aesthetics Implemented protected routes requiring authentication Built responsive forms with real-time validation and loading states Backend (Node.js + Express) Built RESTful API with Express.js Connected to MongoDB Atlas for cloud database storage Integrated Google Gemini AI (gemini-2.5-flash-lite) for intelligent matching Implemented JWT-based authentication for secure login sessions Created comprehensive error handling and validation middleware Used Joi schemas for input validation Built database seeding scripts for colleges, faculties, and majors Database (MongoDB) Designed schemas for Users, Colleges, Faculties, and Matches Populated database with 21 Canadian universities Created relationships between colleges, faculties, and majors Implemented indexing for efficient queries Challenges we ran into AI Integration Integrating Google Gemini AI and crafting the right prompts to get consistent, accurate JSON responses required extensive testing and prompt engineering. We had to balance providing enough context without overwhelming the model.

Authentication Flow Building a secure authentication system with JWT tokens, protected routes, and session management across frontend and backend required careful coordination. Handling token expiration and unauthorized access gracefully was challenging.

Dynamic Form Logic Creating a registration form where majors dynamically load based on the selected faculty required coordinating multiple API calls and managing asynchronous state updates without causing unnecessary re-renders.

Database Design Designing relationships between users, colleges, faculties, and matches while maintaining data integrity and ensuring efficient queries took careful planning and iteration.

Error Handling Implementing comprehensive error handling across the entire stack—from validation errors to network failures to AI service errors—while providing helpful user feedback was complex but essential.