FarmSense
Offline-first coffee lot passport for Tanzanian smallholders: harvest-to-sale traceability, water audits and drying-bed risk alerts.
Offline-first coffee traceability for smallholder farmers, washing stations, and buyers in Tanzania's Mbeya Highlands.
One farmer (Noor), one lot, one drying area: every movement from harvest to sale is recorded as an append-only event and rendered as a buyer-ready passport.
Field workflows use large touch targets and plain language: a farmer home with bed occupancy and a needs-action banner, scan and move coffee between beds, moisture checks against a 10 to 12% target, and a drying-bed guard that turns weather and moisture into a LOW/MEDIUM/HIGH risk with a recommendation. The AI grading and auditing features are simulated in this prototype and are labelled as such in the UI.
Built with Lovable, TanStack Start, React 19, TypeScript, Tailwind CSS 4 and Supabase.