AI Diagnostic
Snap a leaf. Save the harvest.
A computer vision model trained on 40+ crop diseases and pests common in Sahel and tropical climates. Runs on-device, works offline, returns actionable recommendations in the user's language.
Key features
40+ pathologies covered
Mildew, rust, blight, aphids, locust damage, nutrient deficiencies, and more — fine-tuned on field-collected data.
95% field accuracy
Measured on a held-out dataset of 5,000 photos from Mauritanian smallholders.
Offline-first
Model runs locally. Results sync to the dashboard when connectivity returns.
Multilingual treatment plans
Each diagnosis returns step-by-step recommendations in French, Arabic, or English — with safe organic alternatives.
How it works
1. Take a photo
Open the app, tap the leaf or affected zone.
2. Get the diagnosis
Disease + confidence score in under 2 seconds, on-device.
3. Follow the plan
Receive prioritised actions, dosages, and follow-up checks.
4. Share with experts
Forward results to your agricultural agent or veterinarian directly from the app.
Bring Aginov to your country.
We work with ministries, cooperatives, NGOs and impact funds to deploy Aginov in new territories. A pilot can be live in 8 weeks.