Inspiration
We wanted to solve the problem of generic travel plans by creating AI-based personalized itineraries based on individual tastes in food, music, and activities
What it does
TravelItinerary generates a personalized one-day cultural itinerary based on a user's travel destination, interests, food preferences, and group type. It combines cultural recommendations with AI to plan meaningful trips.
How we built it
Frontend: React with Vite, styled using React Hook Form and Zod Backend: Flask API to process inputs and handle requests APIs Used: Qloo for taste-based cultural data, OpenRouter GPT for itinerary generation
Challenges we ran into
CORS errors between frontend and backend Mapping user preferences into clean prompts Handling large frontend file uploads to GitHub Ensuring consistent API responses
Accomplishments that we're proud of
Successfully integrated GPT and Qloo to co-create a dynamic travel planner Created a fully functional full-stack app within the hackathon timeline Built an intuitive UI for non-technical users
What we learned
Prompt engineering for travel and cultural AI Working with real-world APIs and handling response data End-to-end full-stack integration and debugging
What's next for TravelItinerary
Add multi-day itinerary planning Include real-time weather and local events Deploy the app for public use
Built With
- axios-apis:-openrouter-(gpt-based-llm)
- css-frameworks:-react-(vite)
- dotenv-platforms:-localhost-(react-on-port-5173
- flask
- flask-libraries:-react-hook-form
- html
- javascript
- on
- port
- postman
- python
- qloo-taste-ai?-tools:-github
- typescript
- vs-code
- zod
Log in or sign up for Devpost to join the conversation.