ResQ (Team-Berserk)

• 

  Interface

Inspiration

Disasters often escalate faster than information spreads. During floods, cyclones, and earthquakes, people rely on scattered news sources, social media rumors, and delayed government alerts. This gap between real danger and reliable information can cost lives.

We built ResQ to bridge that gap — a real-time disaster intelligence system that combines risk scoring, verified alerts, AI-powered assistance, and emergency tools into a single mobile experience.

Our goal was simple: Make disaster response proactive instead of reactive.

What it does

ResQ is a real-time disaster intelligence and emergency response platform.

It:

Calculates a Universal Disaster Risk Index (UDRI) for a user’s location

Aggregates and verifies disaster alerts

Shows nearest shelters dynamically

Provides an AI-powered disaster safety assistant

Includes an SOS distress system with GPS sharing

Works partially offline with cached survival guides

Automatically triggers periodic safety check-ins during high-risk situations

UDRI=f(weather,source agreement,verification, escalation score) The result is a clear, understandable LOW / MODERATE / HIGH / CRITICAL risk level.

How we built it

ResQ consists of:

Backend FastAPI (Python)

Risk scoring engine

Alert aggregation system

Shelter proximity calculation

Chat API endpoint

SOS distress endpoint

Frontend Custom mobile UI (HTML/CSS/JS)

Risk gauge visualization

Dynamic alert cards

Pop-up survival guides

Chat assistant interface

Cloud Deployment GitHub for version control

Render for backend deployment

The system fetches location-based alerts, calculates risk scores, and updates the UI in real-time.

Challenges we ran into

Handling backend deployment issues (Python version conflicts, build errors)

Preventing API keys from being exposed in GitHub (secret scanning blocks)

Designing a mobile UI that feels native without using React or Flutter

Managing pop-up overlays without breaking layout

Ensuring the app gracefully handles offline states

Deploying a FastAPI app on a free cloud tier with strict constraints

Accomplishments that we're proud of

Successfully built a full-stack disaster intelligence system

Designed a professional mobile-style UI from scratch

Integrated AI-based disaster assistance

Created a real-time risk index (UDRI)

Implemented a GPS-based distress alert system

Deployed the backend live to the cloud

Most importantly, we built something that could genuinely help people during emergencies.

What we learned

Cloud deployment requires strict version control (Python, dependencies)

Security (API keys, environment variables) is critical

UI layering (z-index, overlays, modals) can make or break UX

Risk modeling requires balancing multiple confidence signals

Handling failure states (offline mode) is as important as success states

We also learned how to collaborate efficiently under pressure.

What's next for ResQ

Push notifications for live alerts

Integration with official disaster management APIs

SMS-based fallback system for rural users

Machine learning-based risk prediction

Android/iOS native app release

Multi-language support expansion

Real-time map visualization

Our vision is to make ResQ a globally deployable disaster intelligence system.

Built With

• apscheduler
• css
• fastapi
• github
• google-generative-ai-api
• html
• javascript
• openweather-api
• python
• render
• rest
• uvicorn