CampusMatch

CampusMatch — AI-powered campus dating and networking. From voice intros to venue suggestions, it helps students connect through shared interests and real experiences.

Comment

💡 Inspiration

Campus life is full of opportunities, yet it’s often hard to meet people who share your interests. We wanted to make finding genuine connections — romantic, social, or professional — easier and more natural. That’s why we built CampusMatch, an AI-powered, campus-verified platform that connects students through shared experiences, conversations, and real-world meetups.

💬 What it does

CampusMatch verifies students through university email and voice-based onboarding. The AI transcribes and understands a student’s self-introduction to create a profile automatically. In Phase 1, it recommends campus clubs, events, and venues based on profile data. In Phase 2, the AI analyzes chat topics to suggest shared activities or nearby meet-up spots — for example, recommending a bubble tea café when two users chat about boba. It also syncs with Google Calendar to help users find mutual availability and schedule meetups effortlessly.

🛠️ How we built it

Frontend: React with responsive design for web and mobile.

Backend: Node.js + Express APIs with PostgreSQL for structured data.

AI Integration: Claude API for chat analysis and icebreaker generation.

Voice Onboarding: Speech-to-text pipeline using Web Speech API to capture introductions.

Recommendations: Context-aware engine using Google Places API and interest matching.

Calendar Scheduling: Google Calendar API for automatic time slot suggestions.

Collaboration: Managed through Git with coordinated branching and merging.

⚡ Challenges we ran into

Managing real-time AI chat analysis while maintaining low latency.

Achieving accurate voice transcription from varied accents and noisy environments.

Handling secure OAuth flows for Google Calendar without complicating UX.

Synchronizing backend updates with a rapidly evolving React frontend.

🏆 Accomplishments that we're proud of

Built a fully functional AI-driven matching and recommendation system.

Successfully integrated Claude for contextual conversation support.

Created voice-based onboarding that turns spoken intros into structured profiles.

Implemented calendar scheduling that respects privacy and consent.

📚 What we learned

How to build AI features that enhance human connection rather than replace it.

The importance of team collaboration and Git discipline in a fast hackathon timeline.

Balancing privacy, usability, and intelligence in an app handling personal data.

The power of combining LLMs with real-world APIs to create personalized experiences.

🚀 What's next for CampusMatch

Launching a personalized event feed powered by embeddings and interest clustering.

Introducing AI safety assistants to promote secure meetups and verify venues.

Expanding beyond campuses to alumni networks and city-based communities.

Refining recommendation models with real user feedback to improve match quality.

Built With

Share this project:

Updates