Our proprietary oracle pipeline converts satellite imagery and field observations into a single auditable confidence score — written immutably On-Chain, gating every financing tranche.
Three independent AI systems feed into a single composite oracle score. Each system targets a different stage of the agricultural cycle. All verification data is hashed and stored on-chain — creating an auditable, tamper-proof record for every tranche decision.
Sentinel-1 C-band synthetic aperture radar detects soil disturbance from tillage activity, penetrating cloud cover year-round. Verified before the first tranche of any crop cycle releases.
Sentinel-1 · C-Band SARSentinel-2 optical imagery tracks vegetation health from emergence through harvest. A predictive model projects yield ahead of the harvest logistics tranche, giving investors early visibility.
Sentinel-2 · NDVI · EVIField agents use an on-device AI application to score body condition of every RFID-tagged animal monthly. Results are verified for location compliance and fraud prevention before each livestock tranche.
RFID · On-Device AI · GPSEvery oracle run follows the same four-step pipeline. The process is fully automated, with human review triggered only when confidence falls below threshold. Each step produces an auditable output written.
Satellite and field data collected per scheduled oracle run
Proprietary models score each verification signal independently
Signals combined into a single composite confidence score
Score written to Solana PDA — tranche releases or holds accordingly
Model architecture, accuracy benchmarks, scoring methodology, fraud prevention design, and oracle configuration are available to qualified investors and technical due diligence teams on request.
Request Technical Documentation →We can walk qualified investors through active satellite verification on a funded plot — confidence scores, on-chain records, and the tranche release decision in real time.
Request a demo →