MY ONLY INTENDED SUBMISSION

Inspiration

Food is deeply personal and cultural. We wanted to create a way for people to discover restaurants and cuisines that truly resonate with their taste profile through engaging storytelling rather than cold recommendations.

What it does

FlavorTalks combines OpenAI’s conversational AI with Qloo’s Taste AI to create personalized culinary narratives. Users describe their mood, memories, or dining desires, and FlavorTalks responds with engaging stories while suggesting perfect restaurants and dishes based on their taste intelligence.

How we built it

Frontend: React with Tailwind for responsive UI
Backend: Python FastAPI for API orchestration
AI Integration: OpenAI GPT-4 for storytelling + Qloo Taste AI for recommendations
Database: PostgreSQL for user preferences and story history

Challenges we ran into

NPM & Tailwind Setup Issues: Spent hours debugging dependency conflicts and PostCSS configuration
Multi-API Orchestration: First time coordinating OpenAI + Qloo APIs with different response formats and rate limits
Cultural Context Preservation: Maintaining sensitivity across conversation threads required extensive prompt engineering
Solo Developer Scope: Managing PM/DS/SWE roles simultaneously with tight time constraints
State Management Complexity: React Context became unwieldy as chat features and cultural preferences scaled

Accomplishments that we're proud of

Successful AI Integration: Built working OpenAI + Qloo system that creates culturally-aware restaurant stories
Cultural Bridge Algorithm: Connects familiar cuisines to new ones through shared flavor profiles
Responsive Chat Interface: WhatsApp-inspired UI with real-time typing indicators and embedded restaurant cards
Budget-Aware Storytelling: AI naturally weaves cost considerations into engaging narratives
Full-Stack Solo Delivery: Complete product from database to deployment as single developer
Production-Ready Demo: Compelling user flows showcasing unique cultural intelligence value

What we learned

AI Prompt Engineering: Cultural sensitivity requires iterative refinement and edge case consideration

API Complexity Scales: Multiple external APIs need robust error handling and fallback strategies
Cultural Design Matters: Inclusive technology requires cultural considerations in every design decision
Users Want Explanations: Beta testers preferred why recommendations fit over simple suggestion lists
Time-Boxing is Critical: Strict feature limits prevented perfectionism from derailing progress
Documentation While Building: Real-time notes proved invaluable for presentation and future development

What's next for FlavorTalks

Key Future Advancement

Cultural food tourism integration: Similar to Airbnb, incorporate an experiences feature where you can experience your food anywhere, as well as different adventures eg. you love empanadas and are going to DR, then try wine tasting and zip-lining where you can try their special empanadas.

Immediate (2-4 weeks):