Skip to content
Research programs

One research agenda, four named programs.

Each program below carries its real status. We won’t call something proven until it has been.

01Proven — relaunching

Diagnosis & Alerts

Built, validated, in the field. Being rebuilt for public relaunch.

What started as a weekend build — could a phone camera tell a farmer what was wrong with a leaf — worked well enough to keep going. It grew past its own scope into a system that also warns farmers when disease is spreading nearby: from “here's what's wrong with your plant” to “here's what's happening in your area.” The outbreak-alert layer was already built and already worked in an earlier version of the product. This program is now being rebuilt for a public relaunch within 6 months, meaningfully better than before, not a refresh.

  • Diagnosis accuracy: figures pending
  • Outbreak-alert results: figures pending
  • Relaunch target: within 6 months
02Planned — next up

Soil & Irrigation

Scoped and sequenced. Research and prototyping ahead.

The next program in sequence. Where Diagnosis & Alerts reads a leaf, Soil & Irrigation reads the ground and the water applied to it — built for the same conditions: patchy connectivity, mixed cropping, low-cost sensing rather than enterprise-grade instrumentation.

  • Status: scoping and early prototyping
  • Figures pending
03Planned — next up

Yield Prediction & Planning

Scoped and sequenced. Research and prototyping ahead.

Diagnosis tells a farmer what's wrong now. Yield Prediction & Planning is about what's ahead — using field-level data to forecast outcomes early enough to act on them, so input and timing decisions are made with foresight instead of guesswork.

  • Status: scoping and early prototyping
  • Figures pending
04Open by design

Beyond That

Deliberately unfixed. The mission stays broad until the next program is real.

We're not filling this slot with a roadmap placeholder for its own sake. Existing tools picked one narrow problem and stopped there — our bet is that agriculture's interlocking problems are one research agenda, not a portfolio of separate apps. What comes after Soil & Irrigation and Yield Prediction & Planning will be decided by what the evidence says the world's most common farm needs next.

  • Status: intentionally undefined

Field data, pilots, or co-publication?