Who Is At analyzes your camera footage in real time with artificial intelligence — recognizing people, cutting down false alarms, and processing everything locally, with no cloud required.
Compatible with existing systems. No cameras or infrastructure to replace.
Who Is At started from a personal need: knowing right away if a stranger is in the house and being alerted immediately, without having to upload photos of your own family to someone else's servers — sometimes even paid ones.
That's where the idea came from — a system that processes everything locally, building a face database that genuinely cuts down false positives.
They record everything, but can't tell a real event from a passing car's headlights. The result is hours of footage nobody ever actually reviews.
The artificial intelligence lives inside Who Is At, not in the hardware. Any budget Wi-Fi camera with FTP support is enough to get started — even entry-level models around €40-50 — with no need for cameras with built-in facial recognition, which tend to be far more expensive and often locked into a single closed ecosystem.
Who Is At runs on any on-premise server meeting these minimum specs: x86_64 quad-core processor, 8 GB RAM, 128 GB of SSD storage, a wired network connection to the cameras, Debian 13 (or a Docker-compatible Linux distribution). No graphics card required: AI processing runs entirely on CPU. See pricing →
Four automatic steps, from raw image to a useful alert.
Existing cameras upload a snapshot to an FTP server every few seconds — the minimum interval the camera supports — instead of continuous video recording. No hardware to replace.
Every image is analyzed in real time using YOLO models to detect objects and people.
The detected face is matched against the directory: authorized, known but unauthorized, or unknown.
Only genuinely relevant events trigger a notification, drastically cutting down false alarms.
Everything you need to move from passive recording to monitoring that actually works for you.
A complete, intuitive timeline of every detected event.
Advanced filters by date, camera, person, and event type.
Separate directories for authorized, known-but-unauthorized, and unknown people.
Automatic notifications with camera, timestamp, and — for unrecognized faces — any similarity to a known face, on all the channels you already use.
Who Is At supports integrations with third-party services including Discord, Slack, Telegram, ntfy and e-mail. All product names, logos and trademarks are the property of their respective owners. Their use on this website is for identification and compatibility purposes only and does not imply any affiliation, endorsement or sponsorship.
If an alert isn't handled, it's automatically forwarded to the next channel or person, among those supported.
An overview of recent events, statistics, and recent alerts.
Recognition sensitivity calibrated individually for each camera.
Turn detection on, off, or schedule it by time slot for individual cameras, without touching the rest of the system.
Areas of interest or exclusion, and fixed-object exclusion, defined individually for each camera.
Assign individual cameras to specific operators, to split event review across multiple people.
* Features available at an additional cost over the base plan.
One product for your home, your building, or your business — or for whoever installs it.
Know right away if there's an unfamiliar face at home, without uploading photos of your loved ones to an external cloud — built for the peace of mind of people who live there, not just for professional installations.
A higher level of security for the building you manage, without installing a new system or handing residents' footage over to an external cloud platform.
Offices, warehouses, manufacturing sites: places where false alarms carry a real cost in time and attention, and where knowing who comes and goes actually matters.
You already have the customers, and you're already selling cameras and maintenance. Who Is At becomes an AI add-on you can offer as a new recurring line item, without reinventing your business model.
Compared with the two most common alternatives on the market.
| Feature | Cloud system | Professional AI surveillance | Who Is At |
|---|---|---|---|
| Recurring cost | High (subscription) | High upfront cost | Low (annual license) |
| Data off-site | ✕ Yes | ✓ No | ✓ Never |
| Requires new cameras | ✓ No | Often yes | ✓ No |
| False alarm reduction | Varies | ✓ Yes | ✓ Yes |
| Compatible with existing system | Partial | ✕ No | ✓ Yes |
| Requires cameras with built-in AI (expensive) | Often yes | ✕ Yes | ✓ No — budget Wi-Fi models are enough |
Discover our partner program: turn the maintenance you already do into a high-margin recurring fee.
Get in touch for a demo, the partner program, or any question — we reply within one business day.