CodeCrop calendars
Season-by-season calendars for major crops — sowing windows, fertiliser stages, irrigation timing and harvest — so the year's plan is visible in advance.
Case study — Urdu AI crop advisor for farmers
Agri Dost AI
A farming advisor that speaks Urdu, works offline in the field, and answers before the problem spreads.
AI ProjectIn productionNext.jsFirebaseFlutter
Overview
About this project
Agri Dost AI is a crop advisory product for Pakistani farmers, built Urdu-first because that is the language the field speaks. It gives crop calendars for the season, answers pest and disease questions, and offers weather-based farming guidance — all paired with an offline-first Flutter app, because mobile signal in a cotton field comes and goes. It was built by MindCrafts AI, the AI division of CodeHubb, for the farmers around its own home district of Sargodha and beyond.
The challenge
The problem it solves
When a pest attacks or leaves curl, a farmer's options are slow: wait days for the agriculture department's man to visit, or guess with whatever pesticide the shop recommends. Advice that does exist online is in English, written for other climates, and useless on a phone with no signal standing in a field. The challenge was to put a knowledgeable companion in the farmer's pocket: guidance in Urdu, specific to his crop and season, available exactly where the problem is — in the field, offline.
What we built
Features that do the work
Pest and disease Q&A in Urdu
Farmers describe what they see — curled leaves, spots, insects — and get likely causes and treatment guidance in plain Urdu, standing in the field itself.
Weather-based farming tips
Guidance adjusts to conditions: when to spray, when to hold irrigation, how to protect a crop ahead of harsh weather.
AI assistant chat in Urdu
Open questions go to the AI assistant, which answers farming questions conversationally in Urdu, drawing on the crop knowledge built into the product.
Offline-first Flutter app
Calendars and advice work without a connection; the app syncs when the farmer returns to coverage — designed for fields, not offices.
Private farmer accounts
Each farmer's crops, questions and history live in his own account, private to him (production upgrade, October 2026).
Technology stack
Next.jsFirebase Auth & FirestoreFlutter appUrdu AI chat
Project journey
How Agri Dost AI got here
- October 2026
Concept & design
The concept was set by a single scene: a farmer in a field, phone in hand, no signal, leaves curling. Urdu-first, offline-first and answer-first all follow from that scene.
- October 2026
Build
Crop calendars, the pest Q&A knowledge base, weather guidance and the Urdu AI chat were built, with the Flutter app carrying the same advisor onto the phone.
- October 2026
Quality pass
The advisor was tested against the questions farmers actually ask — in Urdu, mid-season, with the connectivity switched off.
- 21 October 2026
Production upgrade
Agri Dost AI received the production architecture: real farmer accounts, private per-user data, and a server-side AI engine for genuine answers instead of fixed scripts.
Current status
Where it stands today
Production-ready build complete, with accounts, private data, the server-side AI engine and the offline Flutter app finished. Deployment is in progress.
Next step: Launch on its own subdomain once production hosting and the Firebase project are connected, then the Android APK build for field distribution.
Client words
What the client says
“In our fields the signal comes and goes, so an app that needs internet is useless to me. This one works offline, in Urdu. When the cotton leaves curled this season I checked the pest advice standing right there in the field, instead of waiting two days for the agriculture man to visit.”
Want results like Agri Dost AI?
Tell us about your business on WhatsApp — we will map the fastest route from where you are to a project like this one, with a fixed quote before anything starts.