~70%
of the world's freshwater goes to agriculture
Water scarcity
Agriculture is the largest user of the world's freshwater. As aquifers fall and costs rise, every litre has to be justified, and imprecise irrigation is no longer a margin issue.
Bringinggreenhouseefficiencytothefields.
Every litre justified. Every acre optimised.
LeafShift is an AI optimisation layer for irrigation, and soon fertigation. It runs on the sensors and systems you already own, and turns them into one clear answer each day: irrigate this block, this many millimetres, or hold.
Founded 2026Based in the Netherlands
Shadow mode
Starts as recommendations alongside your current setup, so you stay in control.
Confidence you can see
Built to show how sure it is, so you know when to look closer.
Works with what you have
No replacement of equipment. It reads the sensors and controllers already in your field.
No black box
Every decision is logged with its inputs, so you can always see why.
The challenge
The realities of modern farming.
Three forces are squeezing growers from every side. Dutch greenhouses have spent decades answering them under glass, with tight feedback loops and precise control. LeafShift brings that discipline to the open field.
2×+
fertiliser price swings since 2020
Input volatility
Gas-driven fertiliser prices have swung sharply since 2020. Getting fertigation right, alongside irrigation, is where a large part of that budget is won or lost.
5+
disconnected tools per operation
Fragmented systems
Sensors, pivots, pumps, and climate computers rarely share data. Disconnected systems mean reactive decisions instead of one clear picture.
Figures are sector estimates, not LeafShift results.
The LeafShift edge
How the system decides.
Physics does the deciding. Learned components adjust within bounds, and each one has to earn its place before it runs.
Confidence, not guesses
Built to tell you how sure it is.
Most tools give you a single number. LeafShift is built to put a confidence range around every irrigation call, so you know when to trust it and when to look closer.
Day · Manual mmSystem mm · Δ
Mon
-9%
Tue
-10%
Wed
-28%
Thu
-7%
Fri
-23%
Manual vs. systemlogged, not applied
Shadow-mode auditing
Prove the value before you hand over control.
LeafShift runs alongside your current decisions, logging what it would have done and the difference it would have made, without changing anything. You build a data-backed case for or against automation before you commit.
Auditable by design
Every call comes with its reasoning.
Each recommendation carries the state that produced it: how much water is in the root zone, what the crop is using, where it is in its season, and which threshold it crossed. Your agronomist can review it on a specific number. Aggregated pilot data sharpens the models over time.
Decision logZone · today
06:20
IRRIGATE4.2 mm
soil θ 0.21 · ETc 4.810:45
HOLD·
rain likely · conf ↑15:30
IRRIGATE2.1 mm
VPD high · zone 319:10
HOLD·
target moisture metEvery call logged with its inputs
Shadow-mode pilot
Request a pilot.
We are looking for a small number of growers, open-field or under glass, to run LeafShift in shadow mode. Every request is read personally. Tell us about your operation and we will get back to you personally.
Cohort
2026
Basis
By application
Phase
Shadow-mode first
Your data
Stays yours
Your data stays yours. During a pilot we keep a copy to improve our models; you keep full ownership and access. Prefer email? Write to info@leafshift.net.