Sign inGet started
Crowdsourced vehicle recovery

Vehicle recovery, device health, local alerts.

Herm keeps reports, Raspberry Pi modules, and incoming detections in one calm owner portal.

Open dashboardSee device setup

Active reports

1

Devices online

1

Detections 24h

0

Portal previewLive

Active reports

1

Devices online

1

Live network totals from the current Herm backend, with empty states instead of mock values.
Live feed demo
A product-facing version of what owners will actually watch.

Latest matched sighting

empty

No matched sightings have been recorded yet.

Network detections

24h

0 plate sightings logged in the last 24 hours.

Latest module heartbeat

online

Most recent module check-in was 1 hr ago.

Product demo

Longer, lighter, and still anchored in the real product.

The landing page should feel generous. It should explain the system, show enough interface detail to feel tangible, and leave plenty of white space around everything.

 owner ── report plate
   │
   ├─ herm cloud ── match ── notify
   │
 module ── detect plate
   └─ detect person ── owner alert

Vehicles

Report theft, recover faster.
Register owned vehicles, open a stolen report, and keep its status current without jumping between tools.

Sightings

One calm timeline for detections.
Matched and unmatched sightings land with coordinates, timestamps, confidence, and evidence references.

Devices

Pair Raspberry Pi modules cleanly.
Create a device, generate a setup secret, and watch heartbeats, GPS, and camera health from the portal.

Two systems

The crowdsourced network and the owner’s own module coexist cleanly.
One device can help recover a reported vehicle while also keeping watch over its own parked car. The dashboard needs to express both without feeling overloaded.

Stolen vehicle detection

Plate sightings arrive from the network and match against active reports, ready for push fanout later.

Local suspicious activity

Human detections remain tied to the owner’s own device and stay visible as a separate evidence surface.

Flow

How the portal behaves.

Step 1

An owner reports plate ABC123 as stolen.

Step 2

A nearby Herm module detects the plate and posts coordinates.

Step 3

The backend matches the sighting against active reports.

Step 4

The dashboard updates with the event, source module, and evidence.

Evidence

Product proof without crowding the page.

The rest of the page can stay restrained while still surfacing snapshots, alerts, and device state in ways that feel credible during a demo.

01

Snapshot evidence
Every event can carry a still image reference so the dashboard feels concrete during a demo.

02

Local alerts
Human detections from the owner’s own device stay separate from the crowdsourced recovery feed.

03

Status surface
Online devices, stale heartbeats, and setup secrets remain visible without turning the product noisy.

Ready

Start with the dashboard, then connect the hardware.

The web portal is already the operational center. The next step is connecting the Pi module so the detections and alerts become real.

Open dashboardPair a module