Nursery CCTV monitoring without watching the children

A nursery's riskiest minutes are concentrated and predictable. Pickup, when thirty adults arrive in twenty minutes and staff are matching faces to memory. The gate. The edge of the playground. The cameras cover all of it, but a live feed nobody watches is just a recording of things you'll learn about later.

What Svid watches for in a nursery

These run as checks, optional watchers over the same cameras. You choose which cameras and folders Svid watches; a nursery might point them at the entrance and the playground and nowhere else.

• Child Safety flags a child alone where they shouldn't be, outside the play area or by the gate with no adult in frame. It judges a window of recent frames, so a wander reads as an event, not a single blurry still.
• Weapon Alert watches the entrance for the event everyone hopes never happens, raising the alarm as the frames come in.
• Child Safety Review is a report check for afterwards: when a parent raises a concern, it reads the relevant footage and returns a calm, structured account, without a manager scrubbing hours of video.
• Daily Summary is a report check too: it reads the day and ends it with one email of everything that happened, the nursery's answer to “what did my child do today”.

The price comes from your own setup rather than a tariff, never less than $9 a month, with the arithmetic shown on screen at onboarding across all three watch levels. Checks you add later are priced in the Store from your own usage, and report checks start at a $3 floor.

When a parent asks what happened

The footage is searchable in plain English, so typing “child left the play area alone” jumps straight to the moment. A question that used to mean an evening of scrubbing, or an honest “we don't know”, gets a specific answer in minutes.

Getting started

It works with the cameras already on the wall. Create an account at app.svid.ai, answer four questions, and the price appears with its derivation before you pay. Point Svid at the RTSP feeds from your recorder, or at a folder of exported clips if you'd rather trial it on old footage (a check runs across the whole folder in one action), and the first daily summary lands the same evening. Cloud is the default; nurseries that can't send frames out can run the fully on-prem appliance, where nothing leaves the building.

Does this use facial recognition on children?

No. Svid does no facial recognition and stores no biometric data about anyone, children included. There's no tracking of any kind either: the AI describes what a camera sees and checks conditions on it, and nothing follows a child between rooms or builds a profile of one. We built it that way on purpose.

Can parents get a daily update on what their child did?

The Daily Summary report check reads the day's footage and sends an end-of-day email recapping what happened, the nursery's “what did my child do today”. How you share that with parents is your call; much of its value is that the answer exists at all.

Where is the footage stored, and does it leave our building?

By default Svid runs in the cloud: frames go to the vision model to be described, and the footage stays in your account and is never used beyond it. Nurseries that can't allow that can run the fully on-prem appliance, where local models do the work and nothing leaves the building. Appliances keep 30 days of footage by default, and that's configurable.

Can it help at pickup time?

It can watch the entrance like any other camera, and Child Safety flags a child alone where they shouldn't be, by the gate included. It doesn't identify adults or match them to children, so it doesn't replace your collection-password procedure; it's a prompt for a human to look up at the right moment.