Project 02 · OmniVision
Custom cameras through the home, all inference on a Mac on-site. OmniVision learns who's there and what they're doing — and the moment something is wrong, it acts. Footage never leaves the house.
Private by design
On-device
All inference runs on a Mac Studio in the home.
0 frames
of video ever leave the house.
24/7
continuous, tireless awareness.
Real-time
alerts the moment something matters.
What it understands
Recognises actions — eating, sleeping, sitting, walking, cooking — and tracks how long each one lasts.
Knows the resident, family, and carers apart — and notices someone who shouldn't be there.
Learns the normal day and flags deviations — a skipped meal, a restless night, a missed dose.
Catches a fall or a collapse the instant it happens, and escalates without waiting to be asked.
Flags a door or window left open, a stove left on, a walker left out of reach.
Turns hours of awareness into a calm summary — what happened, and what needs a look.
The system
Step 01 · See
Custom cameras install discreetly through the home — by the door, in the living room, the hall. They only ever stream to the local machine; nothing is sent or recorded to the cloud.
Step 02 · Understand
Every frame is understood on a Mac Studio on-site. The model runs locally, in real time — which is what makes the privacy promise real: raw video never has to leave the house.
Add-on
OmniVision doesn't just watch — it can act. A voice agent calls the resident to check in (“are you alright?”), or calls the carer with exactly what it saw and when. It plugs into the channels you already use — joining your WhatsApp via OpenClaw, as an add-on.
14:02 · fall detected — living roomCALLING CARER
09:15 · no movement since 07:40voice check-in
22:31 · front door open 24 minnotify
18:40 · walker out of reachflag
daily summary → carersent 08:00
voice agent · WhatsApp via OpenClaw (add-on)
$
Bring it home
Independence for them, peace of mind for you — without a single frame of video leaving the house.
Request access →