AI-powered travel planning assistant that helps you discover destinations, plan itineraries, and get personalized travel recommendations through intelligent conversations.
- Docker & Docker Compose
- PHP 8.1+ (for local development without Docker)
- Node.js & npm (for assets)
# Clone and start with Docker
git clone https://github.com/dubscode/travelbot.git
cd travelbot
docker compose up -d
# Install dependencies and setup database
docker compose exec app composer install
docker compose exec app php bin/console doctrine:migrations:migrate
docker compose exec app php bin/console app:seed-destinations --count=10
# Install and build frontend assets
npm install && npm run build
# Access the application
open http://localhost:8000- 🤖 AI-Powered Conversations - Real-time streaming chat with travel expertise
- 🌍 Destination Discovery - Personalized destination recommendations
- 📝 Trip Planning - Multi-conversation travel planning sessions
- 🔐 User Management - Secure authentication and conversation history
- 📱 Responsive Design - Mobile-first interface with accessibility support
- ⚡ Real-time Streaming - Server-Sent Events for live AI responses
The heart of TravelBot - A sophisticated RAG (Retrieval-Augmented Generation) system that delivers personalized travel recommendations through:
- 🔍 Semantic Vector Search - Multi-stage search across destinations, resorts, and amenities using pgvector
- 🎯 Smart Query Analysis - NLP-powered extraction of dates, budget, preferences, and requirements
- 📊 Multi-Criteria Ranking - Intelligent scoring based on similarity, user preferences, popularity, and constraints
- 🔄 Progressive Information Gathering - Dynamic follow-up questions to refine recommendations
- 👤 Personalization Engine - Learns from interactions to improve future recommendations
- 🌐 Context-Aware Responses - Temporal, seasonal, and conversation-stage awareness
📚 Complete RAG Documentation →
Backend: Symfony 7.3 with PHP 8.1+ (8.3 in Docker)
Frontend: Twig templates with Hotwire Turbo & Tailwind CSS 4
Database: PostgreSQL with pgvector extension (v15 local, Neon cloud in production)
AI Integration: Claude AI via AWS Bedrock
Infrastructure: AWS ECS Fargate with CDK deployment
# Generate migration after entity changes
php bin/console make:migration
# Check migration status
php bin/console doctrine:migrations:status
# Run pending migrations
php bin/console doctrine:migrations:migrate
# Seed destinations data
php bin/console app:seed-destinations --clear# Development build with watching
npm run dev
# Production build
npm run build# Start all services
docker compose up -d
# View logs
docker compose logs -f
# Execute commands in containers
docker compose exec app php bin/console <command>
docker compose exec app composer <command>- 🧠 Travel Recommendation RAG - Core recommendation engine with 5-service architecture
- 🔍 Vector Search & AI - pgvector implementation with AWS Bedrock Titan embeddings
Production: https://travelbot.tech
Built with modern web technologies for scalable, AI-powered travel assistance.