Q&A: Ranveer Chandra, Microsoft

A few years ago a study looked at a number of sectors and where they were in digital transformation. “It was clear agriculture was lagging,” says Ranveer Chandra, chief scientist, Microsoft Azure Global. “One of the biggest reasons it has been lagging is because it doesn’t have the right data to drive the transformation.”

With its approach, Microsoft tries to bridge the gap. “We can push the boundaries with initiatives like FarmBeats to drive agriculture’s digital transformation,” he says.

Successful Farming sat down with the computer scientist to learn more about the tech giant’s growing presence in agriculture.

SF: Why is Microsoft becoming more visible in agriculture?

RC: Many of the grower’s decisions are based on guesswork today. By capturing large amounts of data from the farm and then using artificial intelligence (AI) and machine learning to translate that data into insights, Microsoft is building technology that replaces the guesswork. 

To enable data-driven agriculture, a seamless data collection system is needed. Our FarmBeats platform enables seamless data collection from a wide variety of sources, including sensors, satellites, drones, and weather stations. It then leverages machine learning, so partner ag tech companies can turn that data into actionable intelligence for farmers. 

Because farms also don’t have great connectivity, Microsoft is investing in technologies such as TV white space, as well as the Azure IoT Edge for collecting data, which can then be ingested in FarmBeats.

Ultimately, FarmBeats can play an important role in sustainably feeding the world with more food – and more nutritious food. It’s also important to point out we are not an agriculture company, and it is not our area of expertise. Our role is to bring Microsoft’s digital innovation to agriculture. To achieve this, we work with a number of partners, and continually add more partners, to the FarmBeats ecosystem who are building solutions for agriculture on top of our platform.

SF: How is AI going to make ag smarter and farms more profitable?

RC: AI can help enable several scenarios for agriculture, from automation to precision agriculture. It can help fill gaps in data from the farm and also predict some values. With the AI in FarmBeats, you can see the soil moisture values throughout the farm with very few sensors. You can also estimate what the soil moisture will be a few days in advance. Several ag applications can be built on top of this AI-built model. For example, when, where, and how much to irrigate, when to sow the seed, when to fertilize, and many more.  

That said, a lot of the work in ag is based on what data you can get from the farm right now, which is very limited. That also limits the amount of AI you can do, because AI is only as good as your data. If you don’t have good data, your AI models will suffer. 

FarmBeats allows us to gather large amounts of data from the farm and from trusted sources in the cloud. By bringing data together (e.g., satellite, weather, tractor, and sensor data), people can start thinking about how to transform the entire food value chain from production to harvest to store shelves.

SF: How is Microsoft making technology more affordable?

RC: No matter what technology we employ, the overall cost has to be affordable for every farmer. In achieving that, data quality must not be compromised.

For example, if farmers want to get an accurate representation of soil moisture, they need to deploy numerous sensors across an entire field, which can be costly. To gain the most intelligence, FarmBeats determines the best location to place sensors in a field. It then combines the data gathered on the ground with aerial imagery so a farmer can build detailed maps of his or her farm with few sensors.

TV white space radios use available TV channels and don’t need dedicated spectrum. Like WiFi, the radios are inexpensive. Because soil moisture and soil electrical conductivity (EC) sensors are expensive, we also realize many farmers aren’t willing to spend hundreds of dollars on this technology. It led us to ask, “How can we help farmers understand what’s going on in their soil at a lower cost?”

SF BIO 

Hometown: Jamshedpur, India

Background: Ranveer Chandra is the chief scientist at Microsoft Azure Global. His research is part of multiple Microsoft products, including Windows, Visual Studio, and Azure. He leads the FarmBeats, battery, and TV white space research projects at Microsoft. Chandra has published over 90 research papers, and he has been granted over 100 patents. 

Education: His undergraduate degree is from IIT Kharagpur, India, and he has a doctorate from Cornell University.