Xplorer
AI-powered travel discovery and collaborative trip planning platform.
— Overview
The Problem
Travel planning typically requires using multiple disconnected platforms for discovering destinations, planning itineraries, and collaborating with friends.
The Solution
Xplorer provides a unified platform where users can explore destinations, generate AI-powered travel plans, chat with friends in real time, and collaboratively organize trips.
The Impact
- Simplifies travel planning through a unified platform
- Improves collaboration through real-time trip communication
- Provides personalized travel recommendations using AI
— Key Highlights
AI-powered travel itinerary generation using Gemini API
Real-time group chat for collaborative trip planning using WebSockets
Mood-based destination recommendations powered by AI
Integrated Stripe payments for premium travel features
Containerized deployment using Docker with CI/CD pipelines via GitHub Actions
— System Architecture
Client Layer
The frontend is built using Next.js and TypeScript to deliver a fast, responsive user interface with TailwindCSS for modern styling.
API Layer
A Node.js and Express server exposes REST APIs responsible for travel data management, itinerary generation requests, and chat session handling.
Authentication Layer
Firebase Authentication secures user sign-in, session management, and protected routes.
AI Services
Gemini API powers AI itinerary generation and mood-based travel recommendations to personalize user travel planning.
Real-Time Communication
WebSockets enable real-time group chat allowing users to collaborate and communicate while planning trips.
Data Layer
MongoDB stores user data, destinations, and trip information while Prisma ORM manages database interactions.
Infrastructure
The application is containerized using Docker and uses GitHub Actions for automated CI/CD pipelines including build, testing, and deployment.
— Features
AI Travel Itinerary Generation
Users can generate personalized travel itineraries based on destination and preferences using Gemini AI.
Mood-Based Destination Discovery
Users can discover travel destinations tailored to their mood such as adventure, relaxation, or cultural exploration.
Real-Time Group Chat
WebSocket-powered group chat enables travelers to collaborate and discuss trip plans in real time.
Secure Authentication
Firebase Auth ensures secure user authentication and protected access to personal travel plans.
Payments Integration
Stripe integration enables users to access premium travel features and services.
Responsive Travel Interface
Fully responsive design ensures seamless travel planning across desktop, tablet, and mobile devices.
— Engineering Deep Dive
Real-Time Communication Architecture
Implemented WebSocket-based communication to enable low-latency real-time chat between trip participants.
AI Integration
Designed structured prompts and API flows to integrate Gemini AI for itinerary generation and travel recommendations.
Database Management
Used Prisma ORM with MongoDB to maintain a clean schema and efficient querying for trip and user data.
CI/CD Automation
Implemented GitHub Actions pipelines to automate testing, building, and deployment processes.
Containerized Deployment
Docker was used to containerize the application ensuring consistent environments across development and deployment.
— Gallery
— What I Learned
Building a full-stack platform using Next.js, Node.js, and Express
Implementing real-time communication systems using WebSockets
Integrating generative AI APIs for personalized travel recommendations
Designing CI/CD pipelines with GitHub Actions
Containerizing full-stack applications using Docker
— Challenges Faced
Managing real-time WebSocket connections while maintaining application scalability
Designing reliable prompts and response parsing for AI-generated itineraries
Structuring backend APIs to support both AI workflows and real-time chat
Maintaining consistent environments across development and deployment using Docker