Pre-launch · design-partner pilots open for 2027Synthetic defects · edge inspection · NVIDIA stackApply for the pilot

Integrations

Fits the cameras, PLCs and systems you already run

Viseena connects to industrial cameras, PLCs, MES and quality systems through open standards, and runs on the NVIDIA stack from data center to edge.

Interfaces

Cameras
GigE Vision, USB3 Vision
PLC
OPC UA (planned)
Events
Webhooks, MQTT
Datasets
COCO, KITTI
Models
TensorRT
Ecosystem

What Viseena connects to

  1. 01

    NVIDIA

    DGX and HGX for generation and training, Jetson and IGX at the edge, RTX workstations for Twinix.

    Compute
  2. 02

    Industrial cameras

    GigE Vision and USB3 Vision area-scan cameras; line-scan support planned.

    Vision
  3. 03

    PLCs

    OPC UA and digital I/O for divert and stop actions (planned).

    Control
  4. 04

    MES and historians

    Verdicts and defect events to MES; process tags from historians for Pilora.

    Operations
  5. 05

    QMS

    Audit records and nonconformance exports for your quality system.

    Quality
  6. 06

    CAD and PLM

    STEP, USD and OBJ geometry; PLM connectors planned.

    Engineering
NVIDIA stack

Built on NVIDIA from end to end

Omniverse ReplicatorDomain-randomized, auto-labeled synthetic defect scenes
TAO ToolkitTransfer learning and per-SKU adapters
DGX / HGXSynthetic generation and model training
Jetson Orin and IGXRuggedized edge inference at the station
TensorRT and DeepStreamMulti-camera pipelines at line speed
HoloscanLow-latency camera and sensor I/O
NeMo, NIM and TritonThe Pilora copilot, served on-premises or in the cloud
OmniverseLine and station digital twins in Twinix
NVIDIA AI EnterpriseSupported deployment for plant IT and OT
API

Events and resources over REST

Synthetic runs, training jobs, models, stations and verdicts are REST resources. Webhooks and MQTT topics push events to your systems as they happen.

  • REST API with token auth
  • Webhooks and MQTT for verdict and drift events
  • Python SDK (planned)
POST /v1/webhooks
# Subscribe to verdict events from one station (example)
curl -X POST "https://api.viseena.com/v1/webhooks" \
  -H "Authorization: Bearer $VISEENA_TOKEN" \
  -d '{
    "url": "https://mes.example.com/hooks/viseena",
    "events": ["verdict.reject", "verdict.review", "model.drift"],
    "station": "line-a-st2"
  }'

# Event payload (abridged, example values)
{ "event": "verdict.reject", "part": "004812", "class": "scratch", "score": 0.97 }
Data formats

Formats in and out

DataFormatsDirection
Part geometrySTEP, USD, OBJIn
DatasetsCOCO, KITTI, Viseena manifest (JSON)Out, and in for your own real frames
ModelsTensorRT engines; ONNX export (planned)Out to Veridex
Station configYAMLIn and out
EventsJSON over webhooks and MQTTOut
Audit recordsCSV and JSON exportsOut to QMS
Next step

Check your stack against Viseena

Send sample images with your camera, PLC and MES list, and we will map the integration.