Local
Learns to stay quiet
Simple

See what matters matters matters важно matters importa важно matters

Yolky Sense is a local AI filter for home cameras. It recognizes your people, remembers false alarms and tells you only what really matters. Over two months at a real home evening false alarms fell 10×, and when the system names someone a human corrects it almost 3× less often.

×10
fewer evening false alarms
45 vs 476 in month one
8%
name corrections by a human
22% at the end of month one
10
cameras on one Mac Mini
was 8, load is lower
Live data: 2 months, Sep 3 – Nov 3, 2026 ↓
0 events seen · 0 notifications
Осень11:00ясно
Simulated scene
Regular camera · noise
Yolky Sense · signal
Stranger at the door
A cat, a branch, a courier, headlights — it all looks like an alarm One notification, only when it is really needed
Noise

Attention is the scarcest resource in a home

Ordinary cameras detect motion the same way for everything. Working out what matters is a job that quietly gets pushed onto you.

01

One alarm for everything

A cat, a branch in the wind, a courier, a stranger — an ordinary camera reacts to motion the same way. Telling important from unimportant is left to you, every single time.

02

Trust fades fast

The first false alarms are annoying, the next ones get ignored. Within a week people simply turn notifications off — and the camera stays silent exactly when it should have spoken.

03

Watching isn’t knowing

The feed shows everything, but it doesn’t answer “what happened while I was away”. You have to dig through the archive to find out.

The same day, two ways to hear about it

Each dot is one event on the cameras on an average day of a real month. Drag the divider.

Every event is an alert≈100a day
Only what matters≈3a day
MorningDayEveningNight

Average day over 29 days, Sep 4 – Oct 2, 2026, from the live database on the Mac Mini. “Regular camera” is every recorded event; “Yolky Sense” is the notifications that were actually sent. Night is the quietest part of the day.

How it works

Not more data — less noise

Every part of the system exists to subtract noise, not to add it.

Learns to stay quiet

Remembers false alarms in specific places — a swing, tree shadows — and gradually stops reacting to them. The longer the system runs, the quieter it gets.

Evening false alarms · month to month
476→45
Twilight stopped flooding the feed: ten times fewer evening false alarms in month two. False-place memory and the night-ghost filter do their job. Full statistic below ↓

Knows its own

Remembers family members and pets — and bothers you only when someone unfamiliar is in the frame.

Runs locally

Video and recognition are processed on your own hardware. By default nothing leaves the home.

No new app

Notifications arrive in Telegram, which you already use.

To a motion sensor, everyone looks alikeYolky Sense knows its own

Neighbors, couriers, passers-by — an ordinary camera reacts to all of them the same way. Yolky Sense remembers who is who and speaks up only when the person is unfamiliar.

Statistic · two months of live data

Two months at a real home, no days off

From Sep 3 to Nov 3, 2026 the system watched the yard around the clock on one Mac Mini. Below — the combined totals. The deep dive into month two is in the full report.

Recorded over 62 days
6,101

camera events. Not a demo or a mock — the live home database.

month one 3,117 · month two 2,984
17,796
human hints over two months
the system learns from them
37
known now: 17 people, 14 vehicles and gear, 6 animals
10
cameras on one Mac Mini
started with eight
1,133
automated tests before each release
908 at the end of month one
Evening false alarms
mo. 1476
mo. 245
10× quieter
Name corrections
end of mo. 122%
end of mo. 28%
when a name is set, it holds better
Vehicles by silhouette
mo. 175%
mo. 291%
face stays at 95%
Names on its own
peak mo. 182%
avg mo. 270%
more cautious — asks more often

What two months add up to

Same yard, same box, same metrics
Events6,101
Human hints17,796
Days with data62
Cameras in use8 → 10
Product commits420

How often the system is right now

Month two, human-reviewed findings only
Facepeople · reviewed 350
95%
Silhouettevehicles · reviewed 576
91%
Silhouettepeople, counting the Neighbors family · reviewed 4,131
86%
Silhouetteanimals and gear · reviewed 67
85%
Place in framevehicles · reviewed 184
32%
90%+70–90%below 70%
Weak spots — we show them ourselves
Twilight

Evening false alarms: 476 → 45

Month one’s main junk source is mostly gone. Night still holds about 2% of findings — the yard is nearly quiet then.

Vehicles

By place in frame — 32%

By silhouette already 91%. Guessing “this car usually stands here” is still below the bar — fix it or turn it off.

Hardware

10 cameras, tight memory

After the unload the Mini runs more cameras at lower load. Free memory is still thin — the next lever.

Month-two report → Month one Weeks, cameras, pairs and incidents — in the full charts.

Numbers come from the live Mac Mini database, snapshot Nov 3, 2026. Periods: Sep 3–Oct 3 and Oct 3–Nov 3. Accuracy is measured on findings a human reviewed. Public reports omit people’s names and pet nicknames.

Trust

Silence shouldn’t cost you trust

Staying quiet when all is well doesn’t mean we stop being honest with you about where your data is kept.

9:41

3
96
31

Processing on-site

Video and recognition are computed on your own hardware. By default nothing leaves the home. In our home, 10 cameras work this way on a single Mac Mini.

Remote access — straight home

The phone connects to your box over an encrypted HTTPS connection. There is no vendor server in between, and no video is stored along the way.

Your data is yours

You decide what to keep and what to delete. There is no middleman who sees the archive by default.

Yolky Sense
Video is processed at your place — nothing goes to a vendor’s server
Basic events are visible without a subscription
The system remembers false alarms and over time stops reacting to them
Only you can see the archive with your guests’ faces
A typical cloud camera
Video goes to the vendor’s server — you rely on their retention policy
Access to the event history is usually locked behind a paid subscription
The same false alarm repeats endlessly — the model doesn’t remember the place
The archive is, in theory, accessible to the vendor and anyone with access to its servers
Demo

One minute to see the whole system

The live interface on our own home: events, entities, zones and rules, analytics and the 3D property map. If you want to try it yourself, request a temporary demo link — names are replaced, frames are stylized, and access can be revoked at any time.

60 seconds · with sound

We’ll review each request individually. Access is temporary and can be revoked at any time.

Playground

Alerts that don’t need your phone

Not every notification has to be a buzz in your pocket. This is where we try calmer, more playful ways to see what’s happening at home — in the corner of a Mac screen and on a tiny pixel display.

30 seconds · with sound. The footage is rendered from real events.
Yolky Notch · macOS

The notch finally has a job

A small menu-bar app turns the MacBook notch into a live ticker for your cameras. When someone arrives, the notch comes alive and shows who it is. One click opens the event card — the frame, the camera, who’s in it — and a tap on the frame switches to the live view.

Recognized someone you know? Confirm or correct it in one tap, and the system gets smarter. Recent events are a swipe away; a do-not-disturb mode keeps it quiet when you need focus.

Live viewOne-tap labelingEvent queueDo not disturb
Tamagotchi mascot

A quiet screen at home

A notification doesn’t have to live in your phone. Yolka is a pet on a Pixoo 64×64 screen in our home: it walks, eats and sleeps, and when the gate or door camera sees a person, an animal or a car, the screen slides away and shows the visitor — calmly, silently, and only when it matters.

The scenes here loop on repeat: the events are simulated.

Camera · person
Summer day, clear Yolka is walking
FAQ

Check the FAQ before writing

Common technical questions live here. If yours isn’t covered — write to us and we’ll look into it.

What is Yolky Sense?

Yolky Sense is an intelligent video-surveillance system that doesn’t just detect motion — it tries to understand what’s happening around the home. It recognizes familiar people, analyzes events, matches observations and helps tell what really matters from everyday noise.

How is Yolky Sense different from a regular AI camera?

Most smart cameras react to isolated events: a person appeared, a car drove by, motion was detected.

Yolky Sense looks at things in context: who appeared, where they were already seen, how confident the recognition is, and whether the owner should be bothered at all.

Is this a finished product or an experiment?

Yolky Sense is a working independent project. The system runs with real cameras around the clock, regularly gets new features and is checked in everyday conditions.

Right now the focus is on technology and recognition quality, not on mass distribution.

Can I install Yolky Sense at home?

Not yet. The system is developed and tested on its own infrastructure. A public version has not been released.

Who is Yolky Sense for?

Primarily for owners of private homes and properties who want to understand what’s happening around them — without constantly watching cameras or drowning in notifications.

What are the minimum system requirements?

The current system runs on a Mac mini M1 with 8 GB of RAM. That hardware is enough for the project’s live configuration.

Minimum requirements for other scenarios are not defined yet.

Which operating systems are supported?

The main working environment is macOS. Windows and Linux will be considered separately.

Which cameras work?

Yolky Sense works with IP cameras that provide an RTSP video stream. The project uses and tests TP-Link Tapo cameras.

RTSP support is an important requirement, but by itself it does not guarantee compatibility of a specific model.

Can I connect several cameras?

Yes. The system processes multiple video streams and can match observations across cameras.

How many cameras can run at once depends on the hardware and processing settings.

Do I need a powerful computer with a GPU?

Not necessarily. The current setup uses a compact Mac mini on Apple Silicon.

The performance you need depends on the number of cameras and how complex the analysis is.

Do I need constant internet?

The core setup is built around a local server and cameras on the home network. Some features, remote access and notifications may still need internet.

Where is the data stored?

In the current setup the system uses its own infrastructure and local storage. That keeps control over the server and access to the information.

How much power does the system use?

Consumption depends on the hardware and load. The current setup uses an energy-efficient Mac mini designed for continuous operation.

Can Yolky Sense recognize people?

Yes. The system recognizes familiar people, uses previous observations and can suggest the most likely matches.

If confidence is too low, the result can be reviewed and confirmed manually.

Do I need to train the system all the time?

No. Confirming new observations is part of everyday use — a few actions in the interface are enough.

As confirmed data accumulates, the system gets more information for later comparisons.

Can the system make mistakes?

Yes. Lighting, distance, angle and image quality all affect recognition.

Instead of hiding uncertainty, Yolky Sense helps you see why a result looks doubtful — and lets you correct it.

Does it work at night?

Yes. Night observations are part of the core scenarios. Extra checks and matching across cameras help improve reliability.

The result still depends on shooting conditions.

Can the system recognize cars?

Yes. Yolky Sense analyzes vehicles and uses extra cues, including color, to match observations more accurately.

Can I exclude certain areas from detection?

Yes. Detection and exclusion zones are supported. They help focus analysis on the right areas and cut down on unnecessary events.

How do notifications arrive?

The system uses Telegram for alerts. The goal is to report meaningful events — not every bit of detected motion.

Can I control the system remotely?

Remote access is part of the current setup. What you can do depends on the server and network configuration.

Is Yolky Sense fully autonomous?

The system can run a large part of its work on its own infrastructure, but some features may depend on external services and an internet connection.

Who is building Yolky Sense?

It is an independent one-developer project that combines computer vision, artificial intelligence, product design and automation.

Does the project use artificial intelligence?

Yes. AI is used for image analysis, object recognition, result checking and working with accumulated observations.

Is Yolky Sense open source?

The source code is not published yet. Whether to open selected components will be decided as the project develops.

Can I suggest a new feature?

Yes. Useful ideas and usage scenarios are welcome. The main development priority, though, is reliability of what already works and fewer false notifications.

Can Yolky Sense be used in commercial projects?

Right now the system is developed for its own use cases. Commercial applications and partnerships can be discussed individually.

Can I invest in Yolky Sense?

The project is open to discussing strategic partnership and investment. The form of collaboration depends on goals, resources and the intended direction of development.

Where can I follow the project?

News, experiment results and system reports are published in Yolky Sense materials. View the project progress report.

Didn’t find an answer?

Please check the FAQ first for technical questions. Feature ideas, partnership and special cases — through the form.

Contacts

I’ll show you the system live and answer your questions

Alexander SheviakovFounder & Product Designer