📖 About the Project 🌱 Inspiration
Navigating a college campus can be unexpectedly difficult—especially for new students. Finding the correct block, exam hall, or administrative office often requires asking multiple people or relying on incomplete information. At the same time, important academic details such as exam schedules or course-related information are scattered across notices, PDFs, and portals.
This inspired NOVA — an idea to combine AI and campus data into a single conversational assistant that students can interact with naturally, just like chatting with a human helper.
🎯 What I Learned
Building NOVA helped me gain hands-on experience with:
Designing a chat-based user interface similar to modern AI tools
Integrating Google Gemini API for real-time AI responses
Managing application state and component structure in React
Using Tailwind CSS for responsive and scalable UI design
Structuring campus data logically (rooms, blocks, navigation paths)
Handling API rate limits and frontend performance issues
Understanding the importance of security, especially API key handling
I also learned that AI systems are not just about calling an API — they require careful prompt design, context control, and efficient request handling.
🛠️ How the Project Was Built
NOVA was built as a frontend-focused AI application using modern web technologies:
React (Vite) was used for fast development and modular UI components
Tailwind CSS handled styling and layout, enabling a ChatGPT-style interface
Google Gemini API powers the conversational intelligence
Campus-related data (blocks, rooms, paths) is modeled in structured JavaScript objects
Admin functionality allows uploading PDFs to extend the assistant’s knowledge base
Navigation logic is based on graph-style path computation
🚧 Challenges Faced
Some of the main challenges included:
API rate limits in the free Gemini tier, requiring request throttling and cooldown logic
Preventing unnecessary API calls caused by component re-renders
Designing a clean chat UI without relying on heavy UI frameworks
Balancing AI flexibility with accurate, campus-specific responses
Keeping the project simple enough for academic use while still being realistic
Overcoming these challenges improved both my technical understanding and my problem-solving approach.
🌟 Outcome
NOVA successfully demonstrates how AI can be applied to real-world campus problems in a practical and user-friendly way. While currently designed for academic and demonstration purposes, the project lays a strong foundation for future enhancements such as backend integration, secure authentication, and interactive visual maps.
Log in or sign up for Devpost to join the conversation.