Among the many problems that beset Indian agriculture is the absence of quality standards in the proverbial farm-to-fork food chain.

While matters of productivity and better market access for farmers often get precedence, the lack of standards disincentivises innovative farmers who try to produce qualitatively better food — and takes away from the consumer the option of buying better.

Even globally, ascertaining the quality of farm produce is a largely subjective exercise, often involving conjecture. Usually, farm commodities are put though physical parameter tests using touch-and-feel, visual appeal, smell and taste. As a result, the ability of farmers who grow higher quality, or pesticide and chemical-free produce, to command a premium is rather limited.

AgNext, a Chandigarh-based agriculture technology (agritech) start-up, could well be among the first in the world to crack this conundrum in a manner that is scalable even in a vast and disparate agri commodities market such as India. Using computer vision (a field of engineering that deals with computers analysing information using digital images and videos), spectral analytics, internet of things (IoT) and artificial intelligence, AgNext’s devices and technology platform can analyse produce quality in about 30 seconds.

Now, an AI solution to ensure tea quality

Easily deployable, fast

“What we have built is like an MRI scanner for food. Besides analysing the physical parameters, we can test the chemical composition of a crop instantly,” says Taranjeet Singh Bhamra, the 41-year-old founder and CEO of AgNext.

While food testing isn’t new and is employed by large commodity trading and food processing companies, it is a cumbersome and time-consuming process that’s carried out in laboratories.

Bhamra says AgNext has not only managed to make test results instantaneous but has also miniaturised the testing devices to such an extent that they can be deployed easily in the field or at a warehouse.

SourceTrace, AgNext tie up to create tech platform for food chain traceability

For instance, AgNext’s milk quality testing device is barely the size of a small home-office laser printer. With a 50-ml milk sample injected into it, the device, using near-infrared spectroscopy (NiR), can instantly measure fat, solids-not-fat (SNF), protein and six other parameters, besides detecting common adulterants such as vegetable oil, urea, detergent, melanin and starch.

The big commercial breakthrough for the start-up founded in 2016 came this month when it inked a deal worth ₹30 crore a year with one of India’s largest private sector dairy firms to test 320 million litres of milk annually.

The deal is close to three times AgNext’s revenues in 2020.

Besides milk, AgNext’s technology platform can currently analyse crops such as rice, wheat, maize, pulses, oilseeds, tea, spices, and animal feed. However, AgNext is not a device company but a form of subscription service. It is the AI-powered software platform that enables the devices to run instant tests. For AI to work, AgNext says it has built vast datasets for each commodity. For instance, it analysed nearly 35,000 milk samples and a million tea leaves while building the platform.

“This suite provides instant quality analysis of composition and contamination, farmer-wise data for quality produce, managing suppliers by lots, and building business intelligence through quality maps,” says Bhamra.

The global market

Bhamra, an IIM Calcutta alumnus, who studied agriculture technology at IIT-Kharagpur, chucked up a career in consulting and investment banking at KPMG in 2015 to pursue his ambition of starting a business that could help Indian farmers.

After a year of “Bharat darshan” and knocking on the doors of dozens of agricultural universities and research institutions across the country to figure out if he could license some technology innovations to build a business, he settled on the idea of digitising the food chain, thanks to the mentorship of scientists and seed money from Association for Innovation Development of Entrepreneurship in Agriculture (a-IDEA), an incubator run by the Indian Council of Agricultural Research (ICAR).

Bhamra has so far raised close to ₹28 crore from Omnivore, an agri-focussed venture capital firm, and Kalaari Capital.

“A lot of Indian agriculture’s problem is on account of the trust deficit, and most of it occurs during the trade of farm produce. There are no objective quality standards nor any means of rapid and reliable testing. During our research while investing in AgNext, we found that there are very few companies, perhaps two or three at best, that have perfected an instant, miniaturised point-of-transaction testing solution. This is one of them,” says Subhadeep Sanyal, a partner at Omnivore.

“This is a sector that has lacked core innovation and any form of technological disruption at all. We believe Indian agriculture will sooner than later become quality-based. In India alone, the addressable market for quality and standards testing is close to $2 billion. We see AgNext as a global player,” says Mandar Dandekar, Principal, Kalaari Capital.

A strong brew

According to Bhamra, the 2018 tie-up with the Tea Research Association, a research body run by the Tea Board, and more than 600 private sector growers in North-East India, turned into a real world, large-scale proof of AgNext’s technology.

The value of tea is determined by what is called the “fine leaf count” (FLC). A unit of two-leaves-and-a bud is counted as a fine leaf. Higher the number of fine leaves in a batch, better is the price the planter gets. For more than a century FLC has been measured using the Tocklai Ballometric Count (named after Tocklai in Assam where the TRA is headquartered) that involves taking a random sample of a few hundred grams of a massive consignment and manually counting the fine leaf. It is a very time-consuming process and TRA felt that greater accuracy of FLC would help small tea growers, who account for 50 per cent of the production, get better prices and improve the overall quality and profile of India’s tea.

“AgNext’s platform determines the fine leaf count of a batch without any human intervention, reducing the process to seconds and improving the overall accuracy. Already installed at 10 collection centres, this technology brings in transparency within the industry, which benefits the farmers and creates a win-win situation for all stakeholders in the value chain,” says Joydeep Phukan, secretary and principal officer, TRA.

comment COMMENT NOW