FurrowIQ
Field-by-field decisions, made from soil, sky and data.
- Client
- Valecrest Farms
- Year
- 2022
- Duration
- 8 months
- Team size
- 5 engineers
Interactive Demo
A hands-on mock of the product we built — click around.
Demo interface shown in English.
The challenge
Valecrest farmed 3,000 hectares the way it always had: blanket fertiliser and spray rates set by the weakest field, irrigation run on calendar rather than need, and scouting notes kept in a pickup-truck notebook. Input costs had jumped 40% in eighteen months, and the estate’s water abstraction licence faced tightening limits under the local catchment plan. Agronomists visited fortnightly, so problems — a rust outbreak, a dry corner — were typically a week old before anyone acted. Coverage made everything harder: half the fields had no mobile signal, so any tool that needed connectivity to function was dead on arrival. The estate needed decisions made per field, sometimes per hectare, from data it already had but never used.
Our solution
We built FurrowIQ around three data planes: Sentinel-2 NDVI imagery processed on a five-day revisit, LoRaWAN soil-moisture probes reporting hourly, and machinery telemetry ingested over ISOBUS where available. A Python and PostGIS backend turns these into per-hectare prescription maps for variable-rate spraying and irrigation scheduling, pushed to a React Native field app that works fully offline and syncs when the tractor reaches the farmyard’s Wi-Fi. Agronomist findings, photos and treatment records attach to GPS points, building a season-by-season history per field. The whole estate runs on a single Kubernetes cluster at the farm office, with a cloud replica for satellite processing.
Results
Technology stack
Facing a similar challenge?
We've solved this class of problem before. Let's talk about how that experience applies to your situation.
Discuss your projectNext case study
Kestrel Pay →