HyperAir

Delhi Air Quality & Public Health Companion

Delhi consistently ranks among the most polluted cities in the world, yet most residents only see a single vague "AQI number" for the entire city — not what's actually happening in their own neighborhood, or what they should do about it. This project was built to close that gap.

The app pulls live air quality data every 30 minutes from real monitoring stations across Delhi, computes the official Indian AQI (using CPCB's piecewise-linear formula, the same method the government uses), and displays it on an interactive map covering 23 stations across the city. Clicking any station shows its exact AQI, the dominant pollutant behind it, a 24-hour history trend, and clear, actionable health advice — like when to skip outdoor exercise, wear an N95, or keep windows shut.

Tech stack: FastAPI (Python) backend, Elasticsearch for fast time-series storage of readings, and a React + TypeScript frontend with a Leaflet map. Everything runs containerized via Docker Compose. If the live data source is ever slow or unavailable, the app gracefully falls back to a realistic synthetic model — shaped from real seasonal and hourly pollution patterns — so the dashboard never breaks, and every reading is honestly labeled as real or synthetic.

Why it matters: Air pollution is a daily health decision for millions of Delhi residents — whether to send kids to school without a mask, whether it's safe to jog outside, whether to keep a window open. A single city-wide number can't answer that. This project's long-term vision is to push toward true street-level AQI, using spatial interpolation and machine learning across denser sensor networks, so anyone can check the air quality right outside their own door — not just the nearest station kilometers away