Global Disaster Alerts Platform

Real-Time Global Monitoring

This project is a core microservice within a larger humanitarian architecture, focusing on the ingestion, validation, and visualization of global crisis alerts. It implements a robust infrastructure capable of handling sensitive data with precision and scalability.

Backend & Cloud-Native Architecture

• High-Performance API: Backend developed with Hono and TypeScript, focused on speed and strict data typing.
• Flexible Persistence: MongoDB Atlas database for a flexible data schema, ideal for the varied nature of different disaster types (earthquakes, famine, etc.).
• Intelligent Processing: AI integration for image recognition (filtering inappropriate content) and Natural Language Processing (NLP) to classify the sentiment of anonymous citizen reports.

Frontend & UX Focused on Critical Data

• Modern Angular: SPA built with the latest Angular 20 features (Standalone APIs and Signals), ensuring a fluid and reactive user experience.
• Geospatial Visualization: Implementation of an interactive map that uses GeoJSON data to display the exact location of events, allowing advanced filtering by type, date, and geographic location.
• Adaptive Design: Interface styled with Tailwind CSS and DaisyUI, featuring dynamic components that change the alert detail display based on the catastrophe type (e.g., earthquake magnitude vs. famine victims).

DevOps & Automated Deployment

• Orchestration with Kubernetes: Deployment of the entire solution on Google Cloud Platform (GCP) using Kubernetes to manage the high availability of deployments and services.
• Configuration Management: Use of Kubernetes Secrets and ConfigMaps to handle sensitive credentials and environment variables securely and decoupled from the source code.
• Serverless Integration: Automated notifications implemented using serverless functions (Lambda/Azure Functions) triggered by alert update events.

Demo Screenshots