Winner Submission for Qloo Global Hackathon 2025
Shaping the future of intelligence using Qloo's Taste AI™ and LLMs

)

Fyndly is a revolutionary AI-powered cultural discovery platform that transforms how people discover content and connect with others through shared interests. By combining Qloo's Taste AI™ with advanced Large Language Models (LLMs), Fyndly creates personalized experiences that understand users' cultural preferences without compromising privacy.

In today's digital age, people are overwhelmed with content choices but struggle to find truly personalized recommendations that align with their cultural tastes and interests. Traditional recommendation systems rely heavily on personal data and fail to understand the deeper cultural connections between different types of content and experiences.

Key Challenges:

• Generic recommendations that don't understand cultural context
• Privacy concerns with personal data collection
• Difficulty finding like-minded people with similar tastes
• Lack of cross-domain understanding (books ↔ movies ↔ music)
• No insights into personal taste patterns and evolution

Fyndly leverages Qloo's cultural intelligence and LLM capabilities to create a privacy-first platform that:

• Understands Cultural Context: Uses Qloo's graph of cultural preferences to make intelligent cross-domain recommendations
• AI-Powered Conversations: Integrates LLMs for natural, context-aware interactions
• Taste Analysis: Provides deep insights into users' cultural DNA and preference patterns
• Social Discovery: Connects users based on shared cultural interests
• Privacy-First: No personal identifying data required

• Smart Content Discovery: Books, movies, podcasts, TV shows powered by Qloo's Taste AI™
• Cross-Domain Intelligence: Understand connections between different content types
• Randomized Fresh Content: Ensures diverse and surprising recommendations
• Fallback Genre System: Robust recommendation engine that never fails

• Cultural DNA Visualization: Interactive brain-map of user interests
• Personality Insights: LLM-generated analysis of cultural preferences
• Content Evolution: Track how tastes change over time
• Personalized Reports: AI-powered insights about user's unique taste profile

• Natural Conversations: Context-aware chat powered by Azure OpenAI
• Content Guidance: Get personalized recommendations through conversation
• Cultural Expert: Understands nuanced cultural connections
• Multi-Modal Support: Text and visual content recommendations

• Interest-Based Matching: Find like-minded people through cultural preferences
• Swipe Interface: Tinder-style discovery for content and people
• Community Features: Connect with others who share your tastes
• Privacy-Focused: Build connections without sharing personal data

• Responsive Design: Seamless experience across all devices
• Interactive UI: Modern animations and micro-interactions
• Onboarding Flow: Intelligent preference collection system
• Real-time Updates: Live recommendations and social features

Frontend:

• Next.js 14 - React framework with App Router
• Tailwind CSS - Utility-first CSS framework
• Shadcn/ui - Modern component library
• Responsive Design - Mobile-first approach

Backend:

• Next.js API Routes - Server-side API endpoints
• Prisma ORM - Type-safe database client
• PostgreSQL - Robust relational database
• Supabase Auth - Authentication and real-time features

AI & ML:

• Qloo Taste AI™ - Cultural intelligence and recommendations
• Azure OpenAI - GPT-4 for conversational AI and analysis
• Vector Embeddings - User similarity and content matching
• Custom Algorithms - Taste analysis and preference mapping

Deployment:

• Vercel - Frontend deployment and serverless functions
• Supabase Cloud - Database and authentication hosting
• Environment Variables - Secure API key management

Key Integration Points:

1. 🎭 Cultural Preference Mapping

   • Maps user favorites to Qloo-compatible genres
   • Fuzzy matching for content identification
   • Cross-domain interest inference

2. 🔄 Dynamic Recommendation Engine

   • Real-time API calls to Qloo's endpoints
   • Randomization for content discovery
   • Fallback mechanisms for robust performance

3. 📈 Taste Evolution Tracking

   • Historical preference analysis
   • Trend identification using Qloo's cultural graph
   • Personalized insight generation

LLM Use Cases:

1. Conversational Recommendations

   • Natural language queries for content discovery
   • Context-aware responses using Qloo data
   • Personalized explanation of recommendations

2. Taste Analysis

   • Deep personality insights from cultural preferences
   • Trend analysis and future predictions
   • Creative content descriptions and explanations

3. Smart Onboarding

   • Intelligent preference collection
   • Genre inference from favorite content
   • Personalized onboarding experience

Landing Page


🧠 Taste Analysis Dashboard

🤖 AI Chat Assistant

📚 Content Recommendations

👥 Social Discovery

![Social Features](.

📱 Mobile Experience

Responsive design optimized for mobile devices

• Node.js 18.x or later
• npm or yarn package manager
• PostgreSQL database (local or cloud)
• Supabase account for authentication
• Qloo API access keys
• Azure OpenAI API access

1. Clone the repository:

git clone https://github.com/sanjail/fyndly.git
cd fyndly

2. Install dependencies:

npm install
# or
yarn install

3. Set up environment variables: Create a .env.local file in the root directory:

# Database
DATABASE_URL="postgresql://username:password@localhost:5432/fyndly"

# Supabase
NEXT_PUBLIC_SUPABASE_URL="https://your-project.supabase.co"
NEXT_PUBLIC_SUPABASE_ANON_KEY="your-anon-key"
SUPABASE_SERVICE_ROLE_KEY="your-service-role-key"

# Qloo API
QLOO_API_KEY="your-qloo-api-key"
QLOO_BASE_URL="https://api.qloo.com"

# Azure OpenAI
AZURE_OPENAI_API_KEY="your-azure-openai-key"
AZURE_OPENAI_ENDPOINT="https://your-resource.openai.azure.com"
AZURE_OPENAI_API_VERSION="2024-02-15-preview"
AZURE_OPENAI_DEPLOYMENT_NAME="gpt-4-turbo"

# Optional: External Services
AZURE_STORAGE_CONNECTION_STRING="your-storage-connection"

1. Set up PostgreSQL database (local or cloud)

2. Run Prisma migrations:

npx prisma migrate dev

3. Generate Prisma client:

npx prisma generate

4. Seed the database (optional):

npm run seed

1. Create a Supabase project at supabase.com
2. Configure authentication providers (Email, Google, etc.)
3. Set up RLS policies for data security
4. Update environment variables with Supabase credentials

1. Sign up for Qloo API access at qloo.com
2. Obtain API credentials from the developer dashboard
3. Test API connectivity:

npm run test:qloo

1. Create Azure OpenAI resource in Azure Portal
2. Deploy GPT-4 model in your resource
3. Configure API endpoints and keys
4. Test LLM connectivity:

npm run test:openai

1. Start the development server:

npm run dev
# or
yarn dev

2. Open your browser and navigate to http://localhost:3000

3. Create an account and start exploring!

# Run all tests
npm run test

# Run specific test suites
npm run test:api
npm run test:components
npm run test:integration

# Test external integrations
npm run test:qloo
npm run test:openai

# Build the application
npm run build

# Start production server
npm run start

• Next.js App Router - Optimal performance and SEO
• Database Indexing - Optimized queries with Prisma
• API Caching - Redis caching for Qloo responses
• Image Optimization - Next.js automatic image optimization
• Progressive Web App - Offline capabilities and app-like experience

• Error Tracking - Comprehensive error monitoring
• Performance Metrics - Real-time performance tracking
• User Analytics - Privacy-respecting usage analytics

• Novel Integration: Seamless combination of Qloo's cultural intelligence with LLM capabilities
• Taste Analysis: First-of-its-kind AI-powered cultural DNA visualization
• Privacy-First: Demonstrates how to build personalized experiences without compromising user privacy
• Cross-Domain Intelligence: Innovative use of Qloo's graph to connect disparate content types
• Conversational AI: Natural language interface for cultural discovery

• Performance: Sub-100ms API response times with intelligent caching
• Responsive: Pixel-perfect experience across all device sizes
• Scalable: Architecture designed for millions of users
• Secure: Enterprise-grade security and data protection
• Tested: Comprehensive test coverage for reliability

• Modern UI: Cutting-edge design with smooth animations
• Accessible: WCAG 2.1 AA compliant for inclusive experience
• Cultural Sensitivity: Respectful representation of diverse cultural content
• Data Visualization: Innovative ways to represent taste preferences

• Global Expansion: Multi-language support and localized content
• Music Integration: Expand to music recommendations using Qloo
• Dining Discoveries: Restaurant and food recommendations
• ✈Travel Planning: Culturally-aware travel suggestions
• Brand Matching: Fashion and lifestyle brand recommendations

• Advanced AI: Integration with latest LLM models
• Predictive Analytics: Forecast future cultural trends
• Community Features: Enhanced social and collaboration tools
• Mobile Apps: Native iOS and Android applications

Built with ❤️ by:

• SJ - Full Stack Developer & AI Engineer
  • 🔗 LinkedIn
• Abinaya - Full Stack Developer
• 🔗 LinkedIn

##License

This project is licensed under the MIT License - see the LICENSE file for details.

• Qloo Team - For providing the amazing Taste AI™ API and cultural intelligence platform
• Azure OpenAI - For the powerful LLM capabilities
• Next.js Team - For the excellent React framework
• Tailwind CSS - For the utility-first CSS framework
• Supabase - For the backend infrastructure and authentication