Skip to main content
Edge devices join AcuSight from the Devices page: generate a one-time setup command, then run it on the device over SSH. It registers itself and appears in your device list — no networking or firewall setup required.
1

Open Devices and add a device

In the sidebar, open Devices, then click Add Device to generate a setup command.
2

Run the command on the device

SSH into the edge machine and run the command AcuSight gives you.
3

The device comes online

It registers automatically and appears in the device list.

What you need on the device

  • A Linux edge machine (arm64, amd64, or arm) with Docker installed.
  • SSH access to the device, and its IP address.
  • Outbound internet access — the device reaches AcuSight over HTTPS. It doesn’t need a public address, open ports, or firewall changes.

Step 1 — Generate a setup command

Open Devices from the sidebar (just below Home) and click Add Device. AcuSight opens Add Physical Device with a ready-to-run setup command:
Add Physical Device dialog

Add Device generates a one-time setup command to run on the edge device.

Click the copy icon to grab the command. It’s single-use and expires after 24 hours — generate a fresh one for each device.

Step 2 — Run the command on the device

Run the command on the edge device itself. SSH in first, then paste it:
The installer pulls the AcuSight agent, starts it, and registers the device against your organization. If the device’s user can’t reach the Docker daemon, run the setup command with sudo.

Step 3 — Confirm it’s online

Back on the Devices page, the new device appears in the list (use Refresh if needed). From there you can point it at a project and start collecting data.

Add a virtual device

Next to Add Device is Add Virtual Device. A virtual device generates simulated device events instead of running on real hardware — handy for trying out projects, dashboards, and event flows before you have a physical camera connected. Create one the same way, straight from the dashboard, with nothing to install.

Next steps

Deploy a model

Push a trained model to your new device or a whole group.

Edge data collection

Point your new device at a project and start streaming data in.

Fleet management

Monitor and manage your devices from the dashboard.

Supported devices

Requirements and the validated hardware platforms.