Spark

Inspiration

Everyone scrolls mindlessly on Instagram Reels, but we have no way of connecting with others who share this specific interest. Social media platforms are good at serving content but quite bad at connecting people based on content you actually enjoy. This sparked our mission: to transform passive scrolling into meaningful connections.

What it does

ReelMatch analyzes the Instagram Reels users have liked to create detailed content preference profiles, then matches users based on compatibility scores. The platform highlights shared interests and suggests conversation starters based on common content preferences.

How we built it

React and Tailwind CSS for the frontend interface Node.js with Express for the backend API MongoDB for flexible data storage A custom matching algorithm combining collaborative filtering with content-based analysis Simulated Instagram API integration (for hackathon purposes)

Challenges we ran into

• Instagram API limitations
• Finding the right balance between content similarity and other factors in our matching algorithm required extensive testing
• Ensuring smooth performance while processing complex matching calculations required creative optimization
• Designing an interface that is easy for the user to understand and doesn't overwhelm them with information ## Accomplishments that we're proud of Successfully developed a functional AI-powered matching algorithm that analyzes content preferences Created an intuitive interface that clearly visualizes compatibility between users Implemented a realistic simulation of Instagram API integration for demonstration purposes Designed an innovative approach to social connections that transforms passive content consumption into active community building ## What we learned Building this taught us many valuable lessons:
• The reels someone likes reveals more about them than a curated profile ever could
• Simple content categorization isn't enough to group together content, context and patterns matter just as much
• Developing user privacy requires a very thoughtful design

What's next for Spark

Securing full Instagram API approval and implementing real user authentication Expanding to additional platforms like TikTok and YouTube Shorts Implementing advanced AI content analysis for deeper preference understanding Building interest-based communities around shared content preferences

Built With

• chatgpt
• claude-ai
• express.js
• mongodb
• react-native
• tailwindcss