Agry OS · Precision agriculture
Agry OS turns field telemetry into decisions and decisions into verified action. Soil, weather and water data arrive over LoRaWAN and 4G, land in one time-series model, and drive irrigation, nutrition and risk management from a single command centre.
Sensors produce readings. Dashboards display them. The gap between a reading and a decision — and between a decision and a verified action in the field — is where yield and water are lost. Agry OS is built to close that gap.
Raw device payloads are stored before decoding, so a corrected decoder can be replayed across history instead of the readings being lost.
A LoRaWAN payload and the equivalent 4G Modbus register resolve to the same measurement. Nothing downstream needs to know which arrived.
Opening a valve is authorised, sent, acknowledged, read back and only then marked verified. An acknowledgement alone never counts as done.
Every recommendation carries its evidence, confidence, model version and the data-quality context it was produced under.
Each stage is a contract the next one can rely on, which is what makes the chain auditable end to end.
Soil probes at three depths, weather stations, flowmeters and valve controllers, over LoRaWAN and 4G.
MQTT uplinks and Modbus polling converge on one telemetry model. Raw frames retained; unknown devices quarantined.
Deterministic thresholds raise alerts with cooldown and auto-resolve — one alert per condition, not one per reading.
Hourly analysis over aggregated data produces recommendations with evidence, expected impact and confidence.
Approved actions become commands that are acknowledged, read back and audited. Models never actuate equipment.
Each module shares the same devices, hierarchy, permissions and audit trail rather than sitting beside the others as a separate tool.
Telemetry, device health, calibration, data quality
Wells, reservoir, pumps, valves, irrigation runs
NPK, EC and salinity, fertigation plans
Rules, approvals, verified command queue
Probability, drivers, inspections, treatments
Weather windows, dosage, compliance records
Forecast, ET, frost, heat and wind risk
Field geometry, infrastructure, NDVI overlays
Cameras, events, visual detections
Water, fertilizer, energy and carbon intensity
Phenology, vigour, yield forecast and harvest
Imaging, grading, defect and batch quality
Executive, manager, agronomist and field views
Scheduling, assignment, evidence of completion
What-if scenarios, attribution, model performance
Tenants, roles, devices, integrations, audit
One component tree serves both directions; layouts use logical properties throughout.
Nine roles across module, action and scope. A user can hold view and edit on water without permission to move a valve.
Every read filters by organisation in SQL, so isolation cannot be forgotten by a caller.
Blocks without sensors are shown as having no data, never as healthy or as an emergency.
Agry OS is deployed at Al Jouf with NADEC. If you run high-tech agribusiness, controlled-environment production or a national food-security programme, we would like to hear what you measure today.
