Voyagr

Voyagr – Turn your trip into a quest. Explore, unlock, and level up your adventure.
User Interface
From preferences to perfect path with Voyagr
Curated Trip Collections
Personalised Routes
Interactive travel quizzes
Leaderboard achievements
The tech behind Voyagr

Inspiration

A couple of us went to Rhode Island and Boston earlier this year. The sites were gorgeous, but the trip was anything from pleasant. We wasted a lot of time trying to figure out where to go, losing out on nearby hidden gems, or taking extensive trips that weren't necessary. But the story of one of our close friends who had an accident and was in a wheelchair at the time made us really examine how we arrange our trips. Things that looked accessible on paper became completely useless in real life: stairs without ramps, uneven paths, and settings without clear entrances. We were tired, frustrated, and overwhelmed, but not because we were exploring. One thing was evident after the trip: travel applications don't work like actual humans do. We needed something that wasn't simply smart, but also knew our limitations, likes, and priorities in real time. That requirement led to the creation of Voyagr. It's not just a route planner; it's a knowledgeable, caring travel companion that helps everyone explore with confidence, no matter what. We want to make travel easy, welcoming, and fun again by using AI, personalized filtering, and routing that takes accessibility into account.

What it does

Voyagr is a smart travel planning application that creates optimized and personalized itineraries in real-time. Users can input their current location and a desired radius, and Voyagr will suggest nearby attractions and activities.

Key functionalities include:

Location-Based Discovery: Instantly find points of interest around you.
Smart Filtering: Customize searches by activity type (e.g., museums, parks, restaurants), specific accessibility requirements (e.g., wheelchair accessible, sensory-friendly), and preferred travel mode (walk, drive, public transport).
Optimized Itineraries: Automatically plans the most efficient route between selected locations, considering travel time and distance.
AI-Generated Content: Leverages Google's Gemini API to provide creative names and engaging descriptions for travel routes and clusters of attractions.
Gamified Experience: Users earn points for visiting locations, with a leaderboard to encourage exploration and add an element of fun.
Clean, Interactive UI: Designed for intuitive use on mobile (React Native Expo), ensuring a smooth and engaging user experience.

How we built it

Voyagr is a full-stack application developed with a focus on modern technologies and efficient data handling.

Frontend: We chose React Native (Expo) to enable cross-platform mobile development, allowing us to build a native-like experience for both iOS and Android users from a single codebase.
Backend: The backend is powered by a Flask (Python) REST API. This handles business logic, data processing, and communication with external services.
Database: MongoDB Atlas serves as our cloud database, storing user preferences, location data, and generated itineraries. pymongo was used for Python integration.

Core Technologies & Algorithms: Route Optimization: We utilized faiss for fast similarity searching and clustering of locations, which helps in identifying relevant points of interest efficiently. This, combined with the Google Routes API, allows for the calculation of optimized travel paths.

   a) AI Content Generation: The Google Gemini Generative AI API was integrated to dynamically 
    create unique and appealing names and descriptions for travel plans.

   b) Location Services: We leveraged a suite of Google Maps APIs:

   c) Geocoding API: To convert addresses to coordinates.

   d) Places API: To search for nearby places.

   e) Custom Search API: For fetching images of locations.

   f) Geolocation API: To determine the user's current location.

   g) Image Handling: Pillow and imagehash were incorporated for image processing and perceptual 
    hashing, which can be useful for identifying similar images or managing visual content related to 
    locations.

Development Environment: We used python-dotenv for managing environment variables and API keys securely. Flask-cors was essential for enabling communication between our frontend and backend.

Challenges we ran into

Building Voyagr in a hackathon environment presented several hurdles:

Juggling multiple Google APIs with varied formats and limits was a complex API integration challenge, which we overcame by developing standardised request handlers and implementing thorough error checking for each API.
Crafting an efficient algorithm for dynamic real-time route optimization was demanding, a hurdle we addressed by leveraging faiss for rapid location clustering and iteratively refining our routing logic based on test cases.
Ensuring consistent data synchronization between the frontend, backend, and database required careful design, which we achieved by establishing clear data models and utilizing asynchronous updates with robust validation.
Achieving a uniform look and feel with cross-platform UI consistency in React Native needed attention to detail, a challenge met by focusing on a component-based design system and conducting extensive testing.
Implementing all desired features within the hackathon's tight time constraints was a constant race, which we managed by prioritizing core functionalities ruthlessly and adopting an agile approach with rapid prototyping.

Accomplishments that we're proud of

Despite the challenges, we achieved several milestones that we're particularly proud of:

Full-Stack Implementation: Successfully developing a complete, functional smart travel planning application from frontend to backend, including database integration.
Meaningful AI Integration: Going beyond simple API calls to use Google Gemini for creative content generation, genuinely enhancing the user experience.
Focus on Accessibility: Integrating smart filters for accessibility needs was a core goal, and we're proud to have built a feature that can make travel more inclusive.
Effective Route Optimization: Implementing a system that provides users with efficient and logical travel plans.
Engaging Gamification: Adding a points and leaderboard system to make the app more interactive and fun.
Intuitive User Interface: Crafting a clean, user-friendly interface that makes complex planning feel simple.

What we learned

The development of Voyagr was a significant learning experience:

Advanced API Utilization: We gained deep practical experience in integrating and managing a diverse set of third-party APIs, particularly from the Google Cloud Platform and Google AI.
Full-Stack Development Workflow: We honed our skills in the end-to-end development process, from conceptualization and design to deployment and testing, using React Native, Flask, and MongoDB.
Importance of Scalable Architecture: We learned about designing backend systems and database schemas that can handle growing amounts of data and user requests.
Problem-Solving Under Pressure: The hackathon environment taught us to think critically, adapt quickly, and find creative solutions to unexpected problems.
Team Collaboration: Effective communication, task delegation, and version control (Git) were crucial for our team's productivity.
The Power of AI in UX: We saw firsthand how AI can be used not just for backend processing but also to directly improve the user-facing aspects of an application.

What's next for Voyagr

We believe Voyagr has the potential to grow into an even more comprehensive travel companion. Some future directions include:

Offline Functionality: Allowing users to access their saved itineraries and essential map data even without an internet connection.
Enhanced Social Features: Enabling users to share their travel plans or completed routes with friends, and perhaps even collaborate on planning.
Community-Sourced Content: Introducing features for users to add reviews, tips, photos, or suggest hidden gems for locations.
Deeper Personalisation: Refining the recommendation engine using machine learning to learn from user behavior and feedback over time, offering even more tailored suggestions.
Expanded Accessibility Options: Adding more granular filters and detailed information regarding accessibility for various needs.
Multi-Modal Trip Planning: Integrating options for planning longer trips that might involve flights or train travel in addition to local exploration.
Improved AI Image Generation: We are currently addressing some challenges with the speed and accuracy of image generation for tours using the Gemini API, and a key next step is to refine this process for a faster and more precise visual experience.

Built With

expo.io
faiss
flask
google-custom-search-api
google-gemini-generative-ai-api
google-geolocation-api
google-maps-geocoding-api
google-places
google-routes-api
imagehash
mongodb-atlas
pillow
pymongo
python
python-dotenv
react-native
supabase

Submitted to

2025 Code For Change: AI Hackathon

Created by

I was responsible for developing the location check-in feature in Voyagr, with a primary focus on backend functionality. To ensure users could only check in at authentic locations, I implemented a perceptual hashing-based image verification system. This technique compared the visual fingerprint of a user-uploaded image with a stored reference image to validate visits. The system was optimized for both accuracy and speed, minimizing false positives while maintaining a seamless user experience. This check-in mechanism became a cornerstone of the app’s location-based authentication, ensuring integrity in user interactions.

In addition to verification, I managed the image storage and route data infrastructure. I integrated Supabase to securely store and serve reference images, taking advantage of its scalability and public URL access capabilities. Simultaneously, I utilized MongoDB Atlas to store and retrieve structured route and location data, supporting dynamic trip management within the app. My backend contributions were essential to making the location check-in process both reliable and scalable, directly supporting the app’s goal of delivering smart, verifiable travel experiences.

Sumit Chugh
Built a comprehensive full-stack route optimization platform combining a custom FAISS clustering algorithm for intelligent location grouping with multi-keyword search and dynamic place discovery capabilities. The system features real-time route computation with adaptive transport modes, AI-powered location naming via Gemini integration, and a sophisticated gamification layer. Advanced functionality includes dynamic hint generation with adaptive difficulty trees, mystery route naming for enhanced engagement, comprehensive optimization algorithms with robust error handling and fallback mechanisms, smart API rate limiting, accessibility filtering, and time-based optimization. Thereby, delivering a complete intelligent routing solution with deep technical sophistication across the entire architecture.

Deekshith Reddy
I was responsible for implementing key user-focused features that enhance the overall functionality and engagement of Voyagr:
• Location Checking: Developed the real-time location-based discovery system that identifies nearby attractions and dynamically adjusts suggestions based on the user’s current position and selected travel radius.
• Profiles Page: Designed and built the user profiles page, allowing travelers to view and manage their activity history, preferences, and progress within the app.
• Leaderboard System: Created an interactive leaderboard that tracks user achievements and rewards points based on exploration milestones, fostering a sense of community and healthy competition among users.

These features were crucial in personalizing the user experience and making travel both accessible and gamified.

RishitRai Rai
I developed the frontend of the application using React Native. This included designing and building screens for trip selection, quizzes, navigation, location verification, etc. I focused on creating a smooth, intuitive user experience using React Navigation and Expo’s camera and location modules.

One of the biggest challenges was implementing the swipe-based feedback feature for the quiz hints. Handling swipe gestures, syncing them with state updates, and ensuring the UI responded in real-time took several iterations to get right. Another major area I contributed to was coordinating with the AI-powered hint generation API. I handled the request/response flow and integrated it with the UI to manage progressive hints based on user feedback.

Throughout the project, I overcame issues like permission handling, multi-screen state flow, and async data rendering , all while ensuring a clean and responsive mobile design. These experiences helped me improve both my frontend engineering and system integration skills.

Pranav Attrey