Qloo Taste Discovery
Inspiration Travel often means losing access to your favorite local spots. We wondered: what if you could find places that feel like your beloved neighborhood cafe, but in Tokyo? Or discover venues with the same cultural vibe as your favorite bookstore, anywhere in the world? Traditional search finds categories - we wanted to find feelings. This inspired us to build a taste discovery platform that maps cultural DNA across global locations.
What it does Qloo Taste Discovery transforms subjective preferences into precise recommendations worldwide. Users input taste descriptions like "cozy minimalist coffee culture" or "vintage bookstore vibes" and receive curated venues that match not just the category, but the entire cultural atmosphere. Our AI understands the deeper connections between places - similar music, crowd, aesthetics, and experiential qualities - delivering personalized discoveries that feel authentically "you" regardless of location.
How we built it Frontend: React/TypeScript with TailwindCSS, featuring an Airbnb-inspired interface with three view modes: infinite-scroll list, interactive split-view, and full map exploration with dynamic pin overlays. Backend: Netlify Functions orchestrating three AI systems: Qloo's Taste AI: Primary cultural similarity engine Google Places API: Real business validation, photos, and ratings
Challenges we ran into Hallucination Prevention: Preventing AI from inventing non-existent places or false connections between user queries and business types. Geographic Accuracy: Ensuring Tokyo queries don't return Poland results, requiring sophisticated location validation and cultural context mapping.
Accomplishments that we're proud of ✅ Validation preventing fake places or incorrect addresses ✅ Optimized architecture delivering fast results worldwide ✅ Professional map integration with interactive pins and adaptive card overlays ✅ Cultural Accuracy: Successfully matching subjective taste preferences to real venues with verified business data
What we learned AI Orchestration: Managing multiple AI systems requires sophisticated timeout strategies, fallback hierarchies, and result validation. User Experience: Subjective preferences need objective validation - users want "places that feel right" but with real photos, ratings, and addresses. Performance vs. Accuracy: Finding the sweet spot between comprehensive search and fast response times through smart batching and parallel processing.
What's next for Qloo Taste Discovery 🚀 Personal Taste Profiles: Learning user preferences over time to improve recommendations 🌍 Social Discovery: Sharing taste maps and following friends' recommendations globally 🎯 Business Intelligence: Analytics dashboard for venues to understand their cultural positioning 💡 Expanded Categories: Beyond dining to include retail, entertainment, and accommodation with the same cultural matching precision
Built With
- google-maps
- netlify
- qloo
- react
- typescript
- vite
Log in or sign up for Devpost to join the conversation.