A safety-first transit companion for international students in Toronto, featuring real crime data visualization, intelligent routing, and multi-modal transit directions.

• Frontend: Next.js 15 (App Router) + TypeScript + Tailwind CSS
• Mapping: Mapbox GL JS + Google Maps APIs + Turf.js
• Authentication: Clerk (with student email verification)
• Database: Supabase (PostgreSQL) + Drizzle ORM
• Font: Geist Sans & Mono
• Deployment: Vercel

• Crime Data Visualization: Real Toronto crime data (June-Aug 2024) processed into risk-level hexagons
• 3D Interactive Maps: Mapbox GL with 3D buildings and smooth navigation
• Risk Level Display: Color-coded hexagons (green=safe, orange=caution, red=avoid)
• Crime Details: Click hexagons to see crime counts, types, and risk scores

• Transit Directions: TTC integration with real departure times and line colors
• Multiple Transport Modes: Driving, walking, cycling, and public transit
• Flexible Scheduling: "Leave at" vs "Arrive by" time preferences
• Route Alternatives: Multiple route options with detailed step-by-step directions
• Transfer Details: Platform information and walking directions between stations

• Natural Search: "Coffee near me" finds nearby places with custom emoji markers
• Instant Routing: Click any place for immediate route calculation
• Google Places Integration: Rich place data with ratings, hours, and photos
• Search Suggestions: Real-time search with Mapbox geocoding

• Modern UI: Dark theme optimized for night use
• Responsive Design: Mobile-first approach for on-the-go use
• Protected Routes: Secure dashboard and profile pages
• Real-time Updates: Live data synchronization

-- Users table with university verification
users (id, email, university, first_name, last_name, created_at)

-- Future expansion tables
safety_reports (id, user_id, location, rating, incident_type, created_at)
buddy_requests (id, requester_id, from_address, to_address, travel_time, status)

git clone <repo-url>
cd pathly
npm install

Create .env.local:

# Clerk Auth
NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY=your_clerk_key
CLERK_SECRET_KEY=your_clerk_secret

# Supabase
NEXT_PUBLIC_SUPABASE_URL=your_supabase_url
NEXT_PUBLIC_SUPABASE_ANON_KEY=your_supabase_key
DATABASE_URL=your_postgres_url

# Mapbox & Google Maps
NEXT_PUBLIC_MAPBOX_ACCESS_TOKEN=your_mapbox_token
GOOGLE_MAPS_API_KEY=your_google_maps_key

npm run db:generate
npm run db:migrate
npm run db:studio # Optional: view database

node scripts/generate-hex-grid.js

npm run dev

• / - Landing page with authentication
• /map - Main mapping interface (protected)
• /dashboard - User dashboard (protected)
• /profile - User profile and settings (protected)

• MapContainer - Main mapping orchestrator
• MapView - Mapbox GL JS integration
• SearchBar - Intelligent search with suggestions
• TransportToggle - Multi-modal transport selection
• DirectionsPanel - Detailed route instructions
• HeatmapControl - Crime data overlay controls

Using Turf.js, we process 50,000+ Toronto crime incidents:

1. Hexagonal Tessellation: 0.2km cells for optimal granularity
2. Risk Scoring: Weighted by crime severity (robbery=1.5x, theft=1.0x)
3. Percentile Distribution: Dynamic risk levels based on data distribution
4. File Optimization: Reduced from 14.8MB to 3.7MB

• POST /api/directions - Google Maps routing with transit support
• POST /api/places - Google Places nearby search
• GET /api/test-db - Database connectivity test

• AI Natural Language Routing: "Get me to campus safely after 10 PM"
• Predictive Risk Modeling: ML-powered safety forecasting
• Buddy Matching: Connect students traveling similar routes
• Real-time Incident Reporting: Community-driven safety updates
• Campus Partnerships: Integration with university safety offices

• International students in Toronto
• Late-night commuters prioritizing safety
• Anyone seeking community-verified route information
• Students unfamiliar with local transit systems

Safety first, always. 🛡️ Made for students, by students. 🎓