TripTeller

Turns your travel photos into vivid, shareable stories. Just upload an image and location, and let our AI craft the narrative for you.

Comment

What inspired us? 💡

In a world saturated with digital photos, our travel memories often get lost in a endless gallery scroll. We capture hundreds of beautiful moments on our trips, but rarely find the time or inspiration to write down the stories behind them. TripTeller was born from this simple problem: What if we could turn those forgotten photos into vivid, shareable stories with just one click? We were inspired to build a tool that not only saves time but also reignites the magic of travel by giving every photo a voice.

What it does 🚀

TripTeller is an AI-powered web application that automatically generates captivating travel narratives from a user's photo.

Here's how it works:

  1. Upload a Photo: A user uploads any image from their travels.
  2. Add Context: They provide the location and choose a narrative tone (e.g., Creative, Funny, Romantic, Horror, or Poetic).
  3. Let AI Do the Magic: Our backend, powered by Google's Gemini 1.5 Flash API, analyzes the image content and context to write a unique, compelling story.
  4. Share Your Story: The user can instantly copy the generated story and share it with the world.

How we built it 🛠️

We developed TripTeller using a robust and efficient tech stack. The backend is built with Python and the Flask micro-framework, making it lightweight and scalable. For the core AI functionality, we integrated the Google Gemini 1.5 Flash API, chosen for its powerful multimodal (text and image) understanding.

The frontend is a clean, modern, and fully responsive single-page application crafted with HTML and styled directly with Tailwind CSS for rapid UI development. The entire application is deployed and publicly accessible via PythonAnywhere.

Challenges we ran into 챌

Our main technical hurdle appeared midway through development. Our initial choice, the gemini-pro-vision model, was suddenly deprecated. This forced us to quickly adapt and pivot. We researched the newly available models and successfully migrated our entire logic to gemini-1.5-flash. This challenge turned into a valuable lesson in adaptability and resilience, strengthening the final product.

What we learned 📚

This hackathon was an incredible learning journey. We went from a simple idea to a fully functional and deployed web application. Key takeaways include:

  • Deeply integrating a powerful multimodal LLM API into a web service.
  • Structuring a Flask application professionally using blueprints and an application factory pattern.
  • The importance of creating a polished and intuitive user experience.
  • How to quickly overcome unexpected technical challenges like API deprecations.

We are proud to present TripTeller as a testament to creativity, rapid development, and the amazing power of modern AI.

Built With

Share this project:

Updates