From farm to fork: a startup serves the food industry a buffet of artificial intelligence

It worked like magic. Computer vision algorithms running in a data center saw that a disease was about to infect a distant cornfield in India.

Sixteen days later, field workers found the first evidence of the outbreak.

It was the kind of magic people like Vinay Indraganti call digital transformation. He’s been practicing it for 25 years, the last dozen of them at companies like Ingredion, a Fortune 500 food ingredients manufacturer.

The project in India was the first major test of AGRi360, a suite of sustainable agriculture products powered by NVIDIA Metropolis from the startup Indraganti co-founded, Blu Cocoon Digital.

The mobile app taps Cloud Smarts

The pilot was both simple and effective.

Farm workers took photos of the plants, time-stamped them and geotagged by a mobile app. They sent them to the Microsoft Azure cloud, where Blu Cocoons’ custom models found patterns that enabled their creepy prediction.

From his background in the industry, Indraganti knows the value of such timely intelligence. You can help farmers and their entire food chain of vendors reap a bumper crop.

It’s a large area, which is why we’ve made AI for food our mantra at Blu Cocoon, he said in an interview from the company’s Calcutta-based suburban Chicago office.

A third eye on the field

AGRi360 acts as a third eye in the field, said Pinaki Bhattacharya, a microbiologist who leads research and development at Blu Cocoon Digital.

In the pilot project, it gave farmers an early warning to apply a small amount of pesticide to halt the disease. An agrochemical company inquired about conditions in the area, helping them manage their supply chain.

In the future, food producers who buy crops will receive key details about their microbiology. This helps plan exactly how and when to process crops into produce to meet the regulatory requirements under which they will be sold.

AGRi360 captures all these insights thanks to artificial intelligence powered by images of farm workers taken as they go about their regular work, said Bhattacharya.

Evaluation of seeds and soils

AI models have begun research by using computer vision to quickly assess soil conditions and seed quality.

These skills are now part of the AGRi360 product portfolio alongside products that monitor plant health and best practices in agriculture. Today, AGRi360 is in use in two countries, improving the quantity and quality of crops.

One customer reports that they are on track to source 100% of their products sustainably by 2025. Another has seen increased revenue for an insecticide, thanks to the service.

Our sales of Cartap 50sp grew 70% in six months thanks to AGRi360’s ability to identify emerging crop infections early, said Vandan Churiwal, director of Krishi Rayasan, a major agrochemicals supplier based in Kolkata.

As a result, we are expanding our license with Blu Cocoon to bring AI-powered insights to every area of ​​our business, he said.

Faster training and inference

Initially, the startup used CPUs to train and run its AI models. It now uses exclusively NVIDIA GPUs and the Metropolis framework for computer vision.

It took two months to train a single AI model on CPUs, Indraganti said. Now, with NVIDIA A10 Tensor Core GPUs, all four models in AGRi360 can be trained within hours, this is a game changer.

The time savings add up quickly because models have to be retrained for new crops, variants, and soil types.

GPUs have also reduced the time it takes to complete inference jobs. Predictions that take 15-20 minutes on CPUs are generated in 2-3 seconds on NVIDIA T4 Tensor Core GPUs. The speed also allows Blu Cocoon to test its models on large and growing datasets.

From shipyards to snack bars

Looking to the future, Blu Cocoon is extending its work in the food chain to container handling in shipyards. He is already testing computer vision models for a customer in India.

We have come up with a way to streamline container handling, reducing their time on site and minimizing touch points to save time and money, said Indraganti.

The startup is even helping food manufacturers create recipes with AI. She’s already made a gluten-free muffin for a packaged food customer with cheese, smoothies, and plant-based snacks next to the menu.

A customer reports that the AI-powered system has helped reduce the time to create a new recipe by 80%.

We named the company Blu Cocoon Digital because we look over the horizon and across the ocean for ways to nurture our customers’ aspirations with digital technology, and it all runs on the NVIDIA platform and Microsoft Azure, he said.

Read Monarch Tractor to discover more ways AI is advancing agriculture.