La Bufalona
A management panel for buffalo dairy farms in Venezuela: daily milking, production history, herd inventory and genealogy, health records and sales, on web and phone, even without signal.
- Sector
- AgTech · Livestock & dairy
- Region
- Venezuela
- Role
- Product design & full-stack development
- Platforms
- Web · iOS · Android
- Year
- 2026
- Stack
- Flutter (Web · iOS · Android)DartBLoC / CubitClean ArchitectureFirestoreFirebase AuthFirebase StorageCloud Functions (TypeScript)Drift (offline-first, WASM on web)go_routerfreezedfl_chart

01
Problem
On a buffalo dairy farm, the numbers that matter (litres milked, cheese produced, yield per animal, which buffalo is in milk, calving, weights, treatments) lived in notebooks and spreadsheets. The owners could not see yield trends, compare periods or trace an animal’s history, and the people recording data work in the field, where the internet comes and goes.
02
Approach
Model the farm’s real day: one daily milking record that feeds every other number, and one record per buffalo that holds its whole life. Then make it work offline first, so a missed signal never means a missed record.
03
Solution
A Flutter app that runs as a web panel and on iOS and Android from one codebase. The dashboard shows animals in milk and waiting, 7-day average yield, and period comparisons (day, week, month, year or custom range). Daily milking (“faena”) captures gross and net milk weight, cheese weight, formula and the people responsible, attaches compressed photos and video of the scale, and calculates litres and yield automatically. Milking history has filters by date and person and a per-buffalo audit. Inventory & Genetics covers the whole herd with RFID search and categories (in milk, dry, calves, heifers), and each buffalo’s record tracks its mother and offspring, calving interval, weight history, individual milk production and medical history (palpation, vaccination, treatments). Sales of milk and cheese compute unit price and compare periods. Roles (super admin, editor, viewer) gate every action, and a Drift database mirrors Firestore so the app keeps reading and writing offline and syncs when the connection returns.
04
Result
The farm runs its daily milking, herd records and sales from one panel, from the office or from the corral, and every litre and every animal now has a traceable history.
Layers I worked on
- Chip & sensors
- Firmware
- Connectivity
- Backend & cloud
- App & web
Source and store links are private for this client.
Screens








