DisasterReady

Before Disaster Strikes, Be DisasterReady

Comment

Inspiration

When our team realized that most disaster-related deaths aren’t caused by the disaster itself but by the lack of early awareness and preparation, we knew something had to change. Nearly 97% of Americans own a mobile phone, yet very few people use it to stay prepared. Disasters feel unpredictable — but with the right data, they don’t have to be.

We asked ourselves: What if survival didn’t depend on luck, but on information you already carry in your pocket? That question became the foundation for DisasterReady.

What it does

DisasterReady is a smart, AI-powered disaster preparedness app that transforms raw alerts into clear, actionable guidance.

It provides:

  • Real-time intelligence using live data from the National Weather Service, FEMA, and OpenWeatherMap
  • AI-based risk predictions up to five days ahead
  • Personalized checklists based on family size, pets, medical needs, and location
  • Smart navigation with Google Maps to nearby shelters, hospitals, and evacuation routes
  • Offline functionality, end-to-end encrypted data, and zero-cost access for all users

In short: It turns panic into action, chaos into coordination, and people into survivors.

How we built it

We built DisasterReady using a layered architecture:

Backend

  • Python FastAPI server

  • Ingests live feeds from:

    • National Weather Service
    • FEMA
    • OpenWeatherMap

  • Processes incoming alerts, normalizes formats, and forwards them to the frontend

  • AI risk-prediction pipeline using pattern matching and historical datasets

  • PostgreSQL database with encryption for secure storage

  • Fallback local caching for offline mode

Frontend

  • React-based mobile interface
  • Clean, emergency-ready UI for quick scanning
  • Tabs for alerts, checklists, preparation plans, and navigation
  • Integration with Google Maps API to route users to shelters and hospitals

Infrastructure

  • Hosted on Vercel + Railway
  • Webhooks to refresh alerts continuously
  • GitHub CI/CD pipeline
  • User-privacy-first design

Challenges we ran into

  • Merging inconsistent data formats from multiple government agencies
  • Ensuring offline functionality despite being a data-heavy app
  • Designing navigation that dynamically re-routes during a disaster
  • Building checklists that automatically update when risk levels change
  • Balancing AI predictions with the need for accuracy and safety
  • UI challenges: making a clean interface that works under stressful conditions

Accomplishments that we're proud of

  • Built a fully working multi-layer disaster-intelligence pipeline in under 48 hours
  • Designed an app that is free, secure, and accessible to all
  • Created an AI model that predicts multi-day risk patterns
  • Built a checklist system that updates in real time with new alerts
  • Implemented offline mode so the app works even after power or service outages
  • Developed a navigation tool that maps nearest shelters, hospitals, and evacuation routes

What we learned

  • How to integrate multiple weather and emergency APIs into one consistent pipeline
  • How to design for low-latency, high-stress usability
  • The importance of accessibility and equity in disaster tech
  • Advanced error handling and offline caching
  • Using AI to support — not replace — verified emergency data
  • How to collaborate effectively under time pressure

What’s next for DisasterReady

We’re just getting started. Upcoming features include:

1. IoT & Smart Home Integration

  • Detect smoke, gas leaks, and earthquakes directly from sensors
  • Trigger automatic alerts and evacuation guidance

2. Wearable Compatibility

  • Push haptic evacuation signals to smartwatches
  • Monitor vitals during disasters

3. Community & Government Integration

  • Collaborate with local governments, schools, and nonprofits
  • Build interactive disaster-education modules
  • Support neighborhood-level response networks

4. Predictive Modeling

  • Improve AI models to anticipate long-range patterns
  • Better routing algorithms under dynamic disaster conditions

5. Global Support

  • Expand beyond the U.S. to support international hazard systems

Our vision: A connected world where every household, school, and community is prepared — not reactive.

Share this project:

Updates