Argus
The Problem
Following geopolitical events through the news is slow and noisy. By the time something shows up in a headline, it’s already been filtered through editorial judgment — shaped by what’s alarming, not what’s structurally significant. I wanted to see raw event data before any of that filtering, organized in a way that makes patterns visible.
What I Set Out to Build
A real-time dashboard that pulls from a massive public database of global events, scores each one by severity, and surfaces escalation patterns as they develop — not after they’ve been packaged into headlines. It needed to handle hundreds of thousands of events per day, show them on a map, and make it easy to spot when a situation is moving fast.
How It Works
Every 15 minutes, the system pulls from GDELT 2.0 — an open public database that catalogues hundreds of thousands of real-world events per day across 100+ countries, sourced from news and media worldwide. Each event gets a severity score, is tagged by type and location, and is written to a local database.
The dashboard presents these as a live, severity-ranked feed. A map view plots events geographically, with size and color reflecting how serious each cluster of activity is. Trend charts show whether a country or region is escalating or de-escalating over time — giving you a data-driven read on where things are moving before it becomes obvious in the news.
The Result
Argus processes 100K+ events per day into a clean, ranked interface. The map view shows global conflict patterns at a glance, and the trend charts surface meaningful developments before they reach mainstream coverage — exactly what the news cycle fails to do.
