Now in Public Beta

Real-Time Geospatial
Intelligence at Any Scale

GPU-accelerated 3D visualization. Sub-millisecond telemetry ingestion. Autonomous operations. One docker pull away.

terminal

$ docker pull registry.tetrapus.com/datamax
$ docker run -it --rm \
    -e DISPLAY=$DISPLAY \
    -v /tmp/.X11-unix:/tmp/.X11-unix \
    --device /dev/dri \
    registry.tetrapus.com/datamax

10,000+

Simultaneous Entities

< 50ns

Send Latency

1,000+

Hz Ingestion Rate

Data at Any Scale

Never lose a data point

The Tetrapus SDK streams data over QUIC with non-blocking sends in ~50ns. Automatic batching, compression, and disk-backed buffering guarantee zero data loss — even through network outages.

QUIC transport with TLS 1.3 and 0-RTT resumption
Disk-backed spill buffer survives crashes and outages
Adaptive bandwidth adjusts to link quality in real time
Priority routing: critical alerts bypass normal batching

main.rs
 use tetrapus_sdk::{TetrapusClient, TetrapusConfig, SchemaId};

let config = TetrapusConfig::new("127.0.0.1:4433")
    .with_queue_capacity(50_000)
    .with_batch_size(1400)
    .with_flush_interval(Duration::from_millis(10));

let client = TetrapusClient::connect(config).await?;

// Non-blocking — pushes to queue in ~50ns
client.send_raw(schema_id, &payload)?; 

Situational Awareness

See your entire operation in real time

A GPU-accelerated 3D globe renders thousands of entities simultaneously with atmospheric lighting, spatial analysis tools, and configurable visual mapping rules.

YAML-driven visual rules: field → colour, size, opacity
H3 hexagonal heatmaps at any zoom level
Spatial analysis: distance, area, range rings, proximity
Camera bookmarks with smooth fly-to animation

visuals.yaml
 # visuals.yaml — map telemetry to visuals
rules:
  altitude_gradient:
    field: "altitude_ft"
    target: "mesh_color"
    gradient: ["#0066ff", "#ff3300"]
    range: [0, 45000]

  fuel_warning:
    field: "fuel_pct"
    target: "marker_scale"
    gradient: [0.5, 2.0]
    range: [10, 100] 

Autonomous Operations

AI that learns from your best operators

TAI watches how your team works and discovers patterns. Claude agents execute approved automations autonomously — every action logged in a hash-chained audit trail.

Pattern mining discovers recurring operator workflows
Human-in-the-loop: operators review before deployment
Claude agents with scoped permissions and tool access
Hash-chained audit trail for full compliance

automation.yaml
 # TAI discovers a pattern:
# "When CO2 > 1000 ppm in Zone A,
#  operator always sets setpoint to 22°C"
#
# Synthesized automation:
- trigger:
    field: "co2_ppm"
    operator: ">"
    threshold: 1000
    zone: "zone_a"
  action:
    command: "set_setpoint"
    value: 22.0
    confidence: 0.94 

Demo