SUV AI
Open the bot
SMART IRRIGATION. LIVING LAND.

Every drop.
At the right
moment.

Day and m³ to irrigate — with a pump, in hours.
Every morning on Telegram — for fields in Uzbekistan.

Open Suv AI Try an example

34 crops · In Uzbek · 06:00 Tashkent time

The advice is built on
your field's data.
34 cropsfrom cotton to orchards
Sentinel + Landsattwo sources of imagery
FAO-56soil water balance
06:00advice on Tashkent time

01 — HOW IT WORKS

From your field.
To a clear decision.

We bring the data together
so it's easier for you to act.

YOUR FIELD ON THE MAPDemo
My field
☀ Sun↝ Wind☂ Rain
➤Advice readyDay + pump hours + volume

It all starts with your plot

Trace the boundary in the Mini App or upload your fields as a file. Choose the crop and irrigation method. The team enters your pump's flow rate when you connect.

02 — FEATURES

Your field.
More clarity.

See what's already available,
what we're testing and where the team can help.

Showing 7 of 14 features

02In the bot

31 crops to choose from

Cotton, wheat, orchards, grapes, vegetables, melons, alfalfa and non-flooded rice. The bot accounts for second-crop corn and summer potatoes automatically.

34 crop profiles in the engine
03In the bot

Your irrigation method

Drip with its wetted fraction, furrow with a pump or by gravity, sprinkler. Flooded rice is not supported.

Drip · Furrow · Sprinkler
04In the bot

Your field by satellite

Sentinel-2, with Landsat 8/9 imagery as backup. NDVI helps assess greenness, NDMI the water content of the vegetation.

Within your boundary, with the image date
05In the bot

Boundaries without the busywork

Trace the field with your finger in the Mini App or import KML, KMZ, GeoJSON or a shapefile — up to 10 fields per chat.

One plot or a whole farm
06In the bot

Everything about the field on one screen

Dala holati brings together irrigation and its uniformity, the weather, the satellite image, spray windows and the reasoning behind the advice.

Dala holati · field screen
07In the bot

Log and costs

Mark “Suv berdim”, rate the advice, and see pump hours and costs once the pump's flow rate is entered. The savings calculation relies on the log entries.

Actual irrigation → next calculation

Features as of 23 September 2026. Access to new tools is expanding gradually.

03 — TRY IT YOURSELF

One question.
A clear action.

Pick a scenario and see
how SUV AI helps through the day.

WHAT DO YOU WANT TO KNOW?

Connect your field ↗

This example is in English.
In the bot, advice comes in Uzbek.

SUV AIDemo example · illustrative data
💧 Suv holati

SAMPLE IRRIGATION ADVICE

Tomorrow — as planned.

In this example the irrigation volume is 160 m³. See how the pump's flow rate changes its running time.

Pump running time2 h
40 m³/h120 m³/h

Time = volume ÷ pump flow rate. The water amount here is illustrative; get the calculation for your own field in the bot.

Example answer, translated✦ From data to action

App · demo

See the whole farm at once.

Map, log and imagery layers. The demo farm is open to everyone; connecting your own fields is limited for now.

Open the app demo ↗

04 — WORKING ON REAL FIELDS

We calculate every day.
We check in the field.

100 haunder the first contract
≈4 hatwo pilot fields: an orchard and a vineyard
862automated tests in the project as of 23 Sep 2026

The first contract is signed — for 100 ha with a large farm. Advice for the pilot has been logged since 9 August 2026. The team: product, engineering and a partner farmer.

01

Images have a date

Clouds and satellite pass times limit observations. Landsat complements Sentinel-2; daily images are not guaranteed.

02

The calculation needs checking

Crop coefficients are FAO-56 table values with an NDVI adjustment. Field inspection, pump data and on-field measurements are needed for local calibration. A satellite index does not replace a soil moisture sensor.

03

Savings are still being measured

There are no measured water savings yet. That requires irrigation records and meter readings; we don't promise savings percentages in advance.

05 — TEAM

The people
behind the advice.

Product, engineering and a partner farmer.
Real names — we stand behind the calculation.

Amir Negmatullaev

Amir Negmatullaev

Author of the idea · product owner

Took SUV AI from the FAO-56 equation to a bot that writes to a real farmer every morning.

Husan Negmatullaev

Husan Negmatullaev

Lead developer

Engine, satellite layer and infrastructure. The physics of the calculation is checked against the FAO-56 guidelines.

Farrukh Urunov

Farrukh Urunov

Pilot partner farmer

Two fields where the product is tested every day. His pump taught the bot to speak in hours, not cubic metres.

06 — SUV AI BLOG

Understand the field.
Decide with confidence.

Explainers on irrigation, satellite observations
and farm economics.

All 9 articles (in Russian)
From practical questions to clear decisions.Choose a topic

A FEW MORE QUESTIONS

Before the first irrigation.

What do I need to start?

Open the bot and answer six short questions: crop, sowing date, area, irrigation method, soil, location. You trace the field boundary later on the Dala holati screen. After irrigating, tap “Suv berdim” — the bot counts from that mark. If the field is on a pump, we enter its flow rate and the advice comes in hours. You don't need to buy sensors to get started; measurements help refine the recommendations.

Which crops and irrigation methods are supported?

The engine has 34 crops: cotton, wheat, orchards, grapes, vegetables, melons, alfalfa and more. Drip, furrow with a pump or by gravity, and sprinkler are available. Rice only on drip or sprinkler; no recommendations are given for flooded rice.

Can I upload several fields at once?

Yes. Besides tracing the boundary in the Mini App, you can import KML, KMZ, GeoJSON and shapefiles — up to 10 fields per chat. For a farm with more fields, the team sets them up.

What happens if the field is under clouds?

Besides Sentinel-2, Landsat 8/9 images are used. If there are no suitable recent images, the calculation relies on the data available. The observation date matters: the satellite doesn't see the field continuously.

How accurate are the recommendations?

A sensitivity test of the calculation, run on live weather for the two pilot fields on 23 August 2026, gives a rough guide of ±1–2 days for “today/tomorrow” advice and ±3–4 days for a week ahead. Without calibration, the volume can be off by up to ±30%. There is no ground truth yet (a moisture sensor or a water meter), so these are the bounds of an internal estimate, not measured accuracy.

Can I already use the app and AI Agronom?

The demo farm in the web app is open to everyone. There's an APK for Android; on iPhone you use the web app. Signing in with your own fields, AI Agronom, weather-based disease risk and alfalfa cut tracking are still in limited demo. The app is not yet published on Google Play or the App Store.

What isn't in the calculation yet?

Groundwater contribution and salt in the irrigation water are not yet taken into account. For fields irrigated with drainage water, the amount may be underestimated — tell us before connecting. The assistant does not replace field inspection or an agronomist's work.

SHALL WE START WITH YOUR FIELD?

Start
with one field.

Meet Suv AI

Add the boundary, crop and irrigation method. Then get advice for your field.