Learn how to leverage geospatial mapping, predictive timelines, and automated alert engines to monitor global threats.
Hypaterra is an advanced geospatial intelligence platform designed for analysts to track, map, and predict global events ranging from geopolitical shifts to epidemiological outbreaks.
The platform consumes data from decentralized intelligence feeds, cleans and standardizes the inputs, and maps them in a unified 3D-globe interface. With built-in analytical dashboards, you can move rapidly from high-level global monitoring to granular statistical analysis.
The **Intelligence Observatory** is the primary real-time monitoring interface.
Events are plotted onto an interactive 3D WebGL globe. Markers are sized and colored based on threat severity (Critical = Red/Pulsing, Moderate = Blue). When clustering occurs in high-density zones, the markers adapt automatically.
Analysts can use the sidebar to filter the globe specifically by region or event class (e.g. "Geopolitical" vs "Health News"). Selecting a marker opens a dedicated detail card providing immediate context about the incident, original source, and timestamp.
The Forecaster shifts the platform from *reactive monitoring* to *proactive prediction* using historical pattern aggregation.
Instead of merely displaying what has happened, the Forecaster analyzes the trailing dataset (e.g., 90 days of records). It builds a threat-calculus based on three distinct matrices:
By overlapping these weights, HypaTerra projects a geographical **Confidence Score (Probability)**.
At the bottom of the Forecaster map, the Timeline Scrubber allows you to fast-forward into the prediction window. Based on historical spread velocity, hotspots will actively pulse, grow, or decay exactly when the algorithm predicts their peak activity day.
Monitoring the globe manually 24/7 is impossible. The Alert Engine allows you to establish autonomous watch-rules.
Navigate to your Operative Profile and access "Active Subscriptions" to draft a new alert. A rule can be as generic as "All Critical Events in Asia" or incredibly granular, such as "Only Economic events in Japan containing the keyword 'inflation'".
When the background ingestion task pulls new intelligence, it parses the metadata against your active rules. A positive match produces an immediate system trigger which is permanently archived and visible inside your Profile dashboard feed.
The HypaTerra Decision Engine is a powerful backend rule-evaluation system. Unlike the simple Alert Engine (which just triggers notifications), the Decision Engine acts as a dynamic logic gateway that assigns confidence scores, calculates risk levels, and generates actionable intelligence recommendations for specific events.
Analysts can create complex logical rules via the Django Admin using the DecisionRule
model. The engine parses JSON conditions to evaluate an event. Here is a breakdown of the critical
JSON keys:
event_type, severity, region,
source).
eq: Exact match.in: The event's value is inside a provided list.contains: The event's value contains the substring (case-sensitive).icontains: Case-insensitive substring match.gt / lt: Greater than or Less than (for numerical
comparisons).risk_level (e.g., "CRITICAL", "HIGH") or a
recommendation string to display in the UI.
When an event is analyzed via the Intelligence Observatory, it is parsed through all active Decision Rules, compiling the triggered protocols into a single, comprehensive situational report.
Often, the fastest intelligence vector during an incident is localized unencrypted FM radio signals.
The Radio Observatory maps over 30,000 global FM endpoints. By switching the globe into Radio mode, analysts can click on emitting sources in target regions and open a direct audio socket streaming the live broadcast, which is crucial for on-the-ground context verification.
The Infrastructure Observatory maps the physical backbone of the global digital economy, correlating hardware residency with organizational dependency.
We track over 1,000 global data centers, including Tier III and Tier IV facilities. Each node is enriched with provider metadata (Equinix, Digital Realty, AWS) and estimated power capacity.
The platform renders the global submarine cable network as animated data vectors. Analysts can identify critical landing stations where trans-oceanic fiber meets terrestrial networks—often the highest-risk chokepoints in the global supply chain.
The Silicon Guard engine is a predictive analytics layer that bridges the gap between software AI models and physical hardware.
Because organizational compute residency is often proprietary, Silicon Guard uses a heuristic engine to infer links based on documented cloud partnerships (e.g., OpenAI/Azure) and geographic proximity to high-capacity hubs.
The engine calculates a Geospatial Connectivity Score (0-100) for any coordinate, evaluating its resilience based on proximity to redundant subsea landings and diversified data center ownership.
The Scenario Sandbox allows analysts to model "What-If" events and observe their propagation through the World Graph.
Simulations use BFS (Breadth-First Search) algorithms to track how a shock at "Ground Zero" travels through entities via relationships like CONTROLS, SUPPLIES, or LOCATED_IN.
The most advanced simulation mode is the Compute Kill Switch. When activated, the engine simulates a total systemic failure of physical infrastructure hubs linked to the target entities. This provides a dramatic visualization of how hardware outages cascade into AI model unavailability and regional digital darkness.