Anya : AI Nutrition Copilot for India

Project-Theme-image

Inspiration

India produces enough food to feed its entire population — yet 35.5% of children under 5 are stunted and 57% of women are anemic. We don't have a food crisis. We have a nutrition intelligence failure.

When we analyzed NFHS-5 data across India's 640+ districts, we found something striking:

In Alirajpur, Madhya Pradesh, 79% of children are anemic — just 800km from Mumbai. These crises are predictable and preventable, but India's nutrition response is always too late, too generic, and culturally disconnected.

The data to solve this already exists — health surveys, food prices, weather patterns, PDS records — but it sits in silos, never connected, never predictive.

So we asked:

What if one platform could predict which deficiencies will spike where, generate precision fortification strategies, and create culturally appropriate meal plans families will actually eat — all within their budget?

That became Anya — meaning "grace" in Sanskrit. Because every child deserves adequate nutrition, regardless of which district they were born in.

What It Does

Anya is an end-to-end AI-powered nutrition intelligence platform built specifically for India. It operates as a three-stage pipeline:

Stage	Purpose	Key Output
PREDICT	Identify who is at risk	Per-district, per-micronutrient risk scores
FORTIFY	Determine what nutrients are needed	Cost-optimized fortification strategies
NOURISH	Deliver meals people will actually eat	7-day culturally aware meal plans

Stage 1 — PREDICT: Who Is At Risk?

Fuses data from NFHS-5, Agmarknet food prices, IMD weather, and PDS records
Generates per-district risk scores for Iron, Vitamin A, Zinc, B12, and Vitamin D
Flags emerging hotspots before crises escalate
Covers India's top 10 most malnourished aspirational districts with real data for 1.6 crore citizens

Stage 2 — FORTIFY: What Nutrients Are Needed?

Generates cost-optimized fortification strategies per district using linear programming
Recommends the right fortification vehicle based on deficiency profile:

Vehicle	Use Case
PDS Rice	Iron and B12 fortification in rice-consuming states
Edible Oil	Vitamin A and D delivery via cooking medium
Double-Fortified Salt	Iron and iodine through universal salt consumption
Mid-Day Meal Flour	Zinc and iron for school-age children

Incorporates bioavailability science — e.g., pairing iron with Vitamin C, separating iron and calcium across meals
Projects 3-month outcomes with estimated reduction in anemia, VAD, and zinc deficiency

Stage 3 — NOURISH: Meals People Will Actually Eat

Generates 7-day culturally aware meal plans based on:

Parameter	Details
Cultural Group	9 groups supported (Tribal Hindu, Muslim, Jain, Sikh, and more)
Season	Adjusted for local and seasonal food availability
Budget	As low as Rs. 2,500 per month
Family Composition	Pregnant mothers, infants, elderly, lactating women

Key capabilities:

Uses only locally available, affordable ingredients including indigenous superfoods (ragi, moringa, drumstick leaves, wild amaranth)
Displays per-meal cost, nutrient coverage, and bioavailability guidance
Includes weekly food group breakdown and key nutrient sources specific to each cultural group

Our model calculates it costs just Rs. 847 per person to shift them out of nutritional risk through this pipeline.

How We Built It

Architecture

Layer	Technology
Frontend	React.js, Mapbox (India district heatmaps), D3.js (charts and visualizations)
Backend	Python, FastAPI
Risk Prediction	XGBoost trained on real NFHS-5 district-level data fused with food price indices, seasonal features, and poverty indicators
Fortification Optimizer	Linear programming via PuLP and SciPy — maximizes deficiency reduction per rupee spent, subject to FSSAI safety limits and bioavailability constraints
Meal Generator	Constraint satisfaction engine combining ICMR Food Composition Tables, WHO/ICMR RDA values, cultural dietary rules (9 groups), local market pricing, and PDS entitlement data
Database	PostgreSQL

Data Sources

Source	What It Provides
NFHS-5	District-level stunting, wasting, anemia rates
Agmarknet	Real-time food prices at mandi (market) level
IMD	Weather and seasonal data
FSSAI	Fortification standards and safety limits
ICMR	Dietary guidelines and food composition tables
Soil Health Card Portal	Regional agricultural data

Everything runs on real government open data — no synthetic or mock datasets.

Challenges We Ran Into

Modeling Indian Food Complexity

A single dish like "dal" varies in ingredients, nutrition, and cost across every state. Building the cultural rules database for 9 religious/cultural groups — with accurate dietary restrictions, pregnancy taboos, and fasting calendars — was the most demanding non-technical challenge.

Data Inconsistency Across Government Sources

NFHS-5 operates at district-level granularity while Agmarknet food prices are reported at mandi (market) level. Reconciling these geographic mismatches required careful spatial mapping and interpolation.

Bioavailability Modeling

Iron absorption varies 3 to 10 times depending on what else is present in the meal. Building the nutrient interaction matrix and enforcing it during meal generation was scientifically demanding:

$$ \text{Absorbed Iron} = \text{Dietary Iron} \times \text{Bioavailability Factor} $$

Where the bioavailability factor ranges from $ 0.03 $ to $ 0.30 $ depending on meal composition (presence of Vitamin C enhancers, absence of calcium and phytate inhibitors).

Multi-Constraint Budget Optimization

Generating a meal plan that simultaneously meets all of the following constraints required solving a complex optimization problem in real-time:

Constraint	Description
Nutritional Adequacy	Must meet ICMR/WHO RDA targets for all micronutrients
Cultural Compliance	Must respect dietary rules for the family's cultural group
Local Availability	Must use only ingredients available in the district and season
Budget Limit	Must fit within Rs. 2,500/month for the entire family
Bioavailability	Must account for nutrient interaction effects across meals

Incomplete District Data

Certain districts had gaps in NFHS-5 coverage. We built fallback estimation logic using neighboring district interpolation and state-level averages weighted by demographic similarity.

What We Learned

Malnutrition in India is not a food problem — it is a data, delivery, and cultural sensitivity problem. The food exists. The government programs exist. What is missing is the intelligence layer connecting prediction to action.
Cultural context is non-negotiable. A nutrition solution that ignores caste, religion, regional food habits, and pregnancy taboos will fail in India, regardless of its technical sophistication.
Indigenous foods are nutritional powerhouses. Red ant chutney (iron-rich), ragi (calcium and iron), moringa (Vitamin A), wild amaranth — these traditional foods outperform expensive supplements but have been systematically overlooked by modern nutrition programs.
Bioavailability matters as much as quantity. A person can consume adequate iron on paper but absorb almost none if it is paired with inhibitors. This invisible factor is a major driver of India's anemia crisis.
Government open data is a goldmine. NFHS-5, Agmarknet, Soil Health Cards, IMD — the infrastructure for intelligent nutrition planning already exists. It simply needs to be connected and operationalized.

What's Next for Anya

Milestone	Description
Full India Coverage	Expand to all 640+ districts with complete NFHS-5 integration and real-time Agmarknet price feeds
POSHAN Tracker Integration	Connect with the government's existing Anganwadi monitoring platform to reach 1.4 million frontline workers
Anganwadi Mobile Application	Offline-first architecture, voice-based UI in 12+ Indian languages, camera-based MUAC measurement for instant child screening
Nandurbar Pilot	Deploy across 50 Anganwadi centers serving approximately 2,000 children in partnership with the district ICDS office
WhatsApp-Based Delivery	Send personalized weekly meal plans to mothers via voice messages in their local language
Dynamic Price Adjustment	Automatically substitute affordable alternatives when local ingredient prices spike
Open Source Release	Release the full platform so other countries facing similar malnutrition challenges can adapt Anya to their context

Social Impact Statement

Anya directly addresses UN Sustainable Development Goal 2 (Zero Hunger) and SDG 3 (Good Health and Well-Being) by targeting the systemic gaps in India's nutrition delivery infrastructure.

The scale of the problem:

Indicator	Statistic	Source
Children under 5 who are stunted	35.5%	NFHS-5
Women aged 15–49 who are anemic	57.0%	NFHS-5
Children under 5 who are wasted	19.3%	NFHS-5
Annual economic loss due to malnutrition	4% of GDP (~\$68 billion)	World Bank

How Anya creates impact:

Predictive intervention, not reactive response. By identifying districts at risk before a crisis peaks, Anya enables targeted resource allocation — reducing waste and improving outcomes.
Cost efficiency at scale. At Rs. 847 per person, Anya's pipeline is significantly more affordable than clinical supplementation programs, making it viable for government deployment across millions of beneficiaries.
Cultural inclusion. By respecting the dietary rules, food preferences, and indigenous knowledge of 9 distinct cultural groups, Anya ensures that nutrition plans are adopted rather than ignored — addressing the single largest failure point of existing programs.
Empowering frontline workers. Integration with POSHAN Tracker and a planned mobile application will put actionable, district-specific intelligence directly in the hands of 1.4 million Anganwadi workers — the backbone of India's nutrition delivery system.
Bridging the data gap. Anya demonstrates that India's existing open government data (NFHS-5, Agmarknet, IMD, FSSAI) can be unified into a coherent intelligence layer — creating a replicable model for data-driven public health intervention.

Projected impact of a single district pilot (Nandurbar, Maharashtra):

Metric	Baseline	Projected After 6 Months
Child anemia prevalence	74.8%	58–62%
Dietary diversity score	3.2 / 9	5.5–6.0 / 9
Monthly nutrition cost per family	Rs. 3,200	Rs. 2,500
Families receiving culturally appropriate plans	0%	100% of pilot group

Anya does not propose a new food program. It proposes the missing intelligence layer that makes India's existing programs — PDS, ICDS, Mid-Day Meal Scheme — work as they were intended to. The food is there. The workers are there. The data is there. Anya connects them.