Evidence, not a mission statement.
This page is a record, not a pitch. Four research programs, each with an honest status — proven, planned, or deliberately open — and the evidence behind it, published as it exists.
Programs
Status matters. We don’t present four programs as equally proven when only one is.
| Program | Status | Summary |
|---|---|---|
| Diagnosis & Alerts | Proven — relaunching | Disease detection from a single smartphone photo, plus regional outbreak warnings when disease is spreading nearby. |
| Soil & Irrigation | Planned — next up | Sensing and guidance for soil condition and water use, spanning software and low-cost hardware. |
| Yield Prediction & Planning | Planned — next up | Forecasting harvest outcomes to help farmers plan inputs, labor, and timing ahead of the season. |
| Beyond That | Open by design | Deliberately not fixed. The mission stays broad until the next program earns a name. |
Our only proven program, held to a published standard.
The current relaunch is being validated against the earlier version’s field results. We’re publishing the placeholders now and will replace them with real figures as the rebuild clears validation — not before.
—
Diagnosis accuracy
Figures pending—
Outbreak-alert precision
Figures pending≤ 6 mo
Time to relaunch
One research agenda, sequenced honestly.
We don’t publish a program until it has a clear thesis and a plan to test it. Diagnosis & Alerts earned its “proven” label from an earlier version that shipped and worked in the field. Soil & Irrigation and Yield Prediction & Planning are next in sequence — scoped, not yet validated, and labeled that way.
Beyond those three, the roadmap is intentionally open. That’s not a gap in planning — it’s a constraint we’re keeping on purpose, so the mission stays broad enough to hold whatever the evidence says the world’s most common farm needs next.