The FAMMO AI Engine
FAMMO's AI engine studies your pet's unique profile and turns it into science-backed nutrition and health guidance. Here's what happens behind the scenes.
What goes in
Your pet's full profile: species, breed, age, weight, activity level, body condition, food preferences, allergies, and known health issues. The more complete the profile, the sharper the recommendation.
{ "species": "dog", "breed": "labrador", "weight_kg": 24.5, "activity_level": "moderate" }
Nutrition calculation
The engine calculates daily calorie needs, ideal protein/fat/carbohydrate ratios, and identifies the right diet style for your pet (like weight loss, senior wellness, or growth) from 10 possible diet profiles.
{ "calories_per_day": 780, "protein_percent": 28, "diet_style": "weight_loss" }
Six-category risk assessment
Every profile is screened across six health risk categories: weight, joint, digestive, metabolic, kidney, and dental. These are preventive indicators, not medical diagnoses, designed to flag what's worth watching.
{ "weight_risk": "high", "joint_risk": "medium", "kidney_risk": "low" }
Personalized meal plans
The AI builds complete meal options with ingredients, portion sizes, a feeding schedule, and safety notes, tailored to your pet's exact targets.
{ "meals_per_day": 2, "portion_size_grams": 195, "safety_notes": ["Avoid onions and garlic"] }
Health reports
A separate AI pass produces a full health report: breed-specific risks, weight and diet guidance, feeding tips, activity recommendations, and alerts.
{ "breed_risks": [...], "activity": "30-45 min daily", "alerts": ["Schedule annual hip X-ray screening"] }
NURNO, your AI companion
NURNO uses the same understanding of your pet to answer questions in real time, with text or photos, any time you need it.
{ "role": "user", "text": "Is this food safe for my senior dog?" }
Always learning
Every prediction is logged and studied to keep improving accuracy over time, as FAMMO grows from more pet profiles and outcomes.
{ "logged": true, "model_version": "1.0.0", "confidence_score": 0.85 }