Pune Startup Froncort.AI Gets $2.3 Mn Google.org Grant For AI Disease Screening App

Froncort.AI, a Pune-based healthcare AI startup, has received a $2.3 million grant from Google.org to build a mobile tool that tells frontline health workers which households to screen first for tuberculosis and other infections.

The grant comes with $200,000 in Google Cloud credits and pro bono engineering support. It is Google.org’s second grant to the company, after funding a validation study of its AI-NETRA platform in 2025.

How the Company Was Selected

Froncort.AI is one of 15 organisations worldwide picked for the Google.org Impact Challenge: AI for Government Innovation, a $30 million programme whose recipients were announced on 15 September.

Read More: Sachin Tendulkar Backs Waste Management Startup Recykal

FundingConfido Health Raises $12.5 Million To Automate Healthcare Administration With AIFundingProdoc AI Raises Seed Funding From Dr A. Velumani Family Office

Google.org said winners were chosen from more than 2,600 proposals. Froncort.AI is one of four recipients in the health category, alongside Flexbone, Helium Health Foundation and the University of Warwick.

Read More: How to Choose the Best Travel Agency in Jaipur for a Memorable Rajasthan Holiday

That ratio explains why the grant matters more than its size. Being selected from 2,600 applicants is third-party validation that a health ministry or a state programme can cite.

What mAI-NETRA Does

The grant funds mAI-NETRA, a mobile version of the company’s public health platform.

Froncort.AI’s system combines government health programme data with satellite-derived insights from Google’s Earth AI models. It also weighs local indicators such as settlement density, land use, rainfall and flood exposure.

From that, it flags higher-risk areas within 300-metre zones and turns them into ranked work plans, household clusters and routes that health workers can follow on field visits.

The Froncort.AI app works offline with automatic sync and offers multilingual voice guidance. Matters in districts where connectivity is patchy.

Why Routing Beats Testing Alone

India’s tuberculosis programme already has tests, drugs and reporting systems. What it lacks is a way to decide where to spend a health worker’s day.

Sending a worker street by street in a high-burden district is expensive and random. Sending the same worker to a ranked cluster of households with recent flooding and dense housing is a different use of the same salary.

Froncort.AI already has TB and malaria modules are already in place. The platform is being built to cover six diseases: TB, leprosy, malaria, dengue, cholera and leptospirosis.

What the Founder Says

“The world does not lack experts. It lacks their reach,” said Ujjwal Rao, founder and chief executive of Froncort.AI. He described AI-NETRA as putting that expertise in the health worker’s hands at the door, showing where to go, whom to screen and when.

India’s health system already collects enormous amounts of data through national programmes. The gap is turning that data into a route a health worker can follow on a Tuesday morning.

Satellite inputs add something the health records cannot: environmental risk. Flood exposure and land use predict leptospirosis and vector-borne outbreaks before a single case is reported.

Froncort.AI’s constraint is field support. A model that guides screening only helps if district officers trust it and if ASHA workers can use it on low-end phones.

The programme is a useful test of whether philanthropy can fund public health AI at the pilot stage, where venture capital rarely goes.

What to Watch

Three things will decide whether this scales. The first is state adoption, because a central pilot that never reaches district workflows changes nothing.

The second is field accuracy. A routing model that sends workers to the wrong households wastes the scarcest resource in the system. Is their time.

The third is data integration. The platform is only as good as the government datasets it can legally and technically connect to. That varies by state.