Web-based immersive wildfire simulation with 3D visualization, AI-powered risk analysis, voice alerts, and evacuation mapping.

• Frontend: React, Vite, TypeScript, Three.js
• Backend: Python, FastAPI
• AI: Google Gemini
• Voice: ElevenLabs TTS
• Media: Cloudinary
• Networking: Tailscale Funnel
• Deployment: Docker, Vultr

cp .env.example backend/.env

docker compose up --build

Backend: cd backend python3 -m venv venv source venv/bin/activate pip install -r requirements.txt uvicorn app.main:app --reload

Frontend (new terminal): cd frontend npm install npm run dev

open -a Tailscale tailscale up tailscale funnel 8000

• GET / Health check
• GET /api/wildfire-risk Quick risk score
• POST /api/analyze-fire Full AI analysis
• POST /api/voice-alert - POST /api/voice-alert - POST /api/voice-alert - POST /api/v##- POST /api/voice-alert - POST /api/voice-alert - POST /api/v - POST /api/ |- POST /api/voice-alert - POST /api/voice-a---- POST /api/voice-alert -| - POST /api/voice - POST /api/voice-aleroo- POST /api/voice- - POST /api/voice-alert - POS h- POST /api/voice-alert -sett- Ps | | CLOUDINARY_CLOUD_NAME | https://console.cloudinary.com | | CLOUDINARY_API_KEY | https://console.cloudinary.com | | CLOUDINARY_API_SECRET | https://console.cloudinary.com |