Bloomfield Robotics Wins Funding to Feed People on Earth and in Space

Anyone embarking on a startup today is entering a business and technological landscape that is vastly different from what it was even two years ago. Cyberattacks have skyrocketed since 2020 — and not only for large organizations. Increasingly, small to medium-sized organizations have been targeted to disastrous effect. 

This new reality was part of Akamai’s drive to create a $1 million challenge that would provide a startup with the power and protection they need to survive and thrive. We were also looking for a company to partner with us in making lives better for billions of people, billions of times a day. 

Improve crop yields: globally and intergalactically

Our extraordinary winner, Bloomfield Robotics, does that and more.  The company aims to improve food security, food quality, and use of arable land to improve crop yields across the globe — and even in space. 

Bloomfield Robotics has a straightforward business model: using portable custom cameras, which can be attached to any farm vehicle, to capture images of individual plants. These images are then uploaded to the cloud, where key plant features are analyzed using deep learning algorithms that determine the health of plants against expectations. Farmers pay an affordable monthly fee for the tools and technology to assess their plants.

A Q&A session with the CEO

I sat down with Mark DeSantis, CEO of Bloomfield Robotics, to talk about the journey his team has taken thus far and what winning $1 million in computing and security solutions means for them.

What was the genesis of Bloomfield’s approach to increasing crop yields?

Startup life is my way of life, so this isn’t my first one. The idea of using images to predict problems or solve them was central to a previous company that I worked with called RoadBotics, which is still alive and well. There, we used image capture from mobile phone cameras to analyze the condition and health of asphalt road surfaces in major cities in the United States and Europe. 

Certain patterns of cracks or deformations, for example, could be cured quickly when identified early, and this would prevent dangerous potholes and expensive repairs down the road.

So, how did you think to apply that to plants?

Well, throughout my career, I’ve been interested in solving big problems with technology. Food insecurity worldwide is one of the biggest. World events, such as weather and war, only make it worse. The Russian war in Ukraine, for example, has already added 40 million more to the number of people experiencing food insecurity. 

So, if all the land that’s ever going to be farmed is being farmed right now, and we will be adding 40% to the world’s population over the next 30 years, each plant we grow is going to have to deliver more fruit, more vegetables. I wanted to figure out how we could do this. 

Then it came to me how my Italian immigrant grandfather grew the most amazing tomatoes in his backyard garden. He was in there every day with his plants, fussing around, inspecting them over and over. He gave each plant constant, individual attention so it would not only produce more fruit, but the best fruit possible. I basically needed to use technology to scale my grandfather. 

My idea was that artificial intelligence could give modern farmers the ability to give every one of their plants individual, unique attention, which could extend the outcomes my grandfather achieved in his backyard to acres upon acres.

How do you photograph plants so they can be analyzed for potential signs of disease or other problems?

We use our FLASH camera, mounted on a Kubota or other field implement, to photograph the plants. At its core, Bloomfield is a repository of crop health data that can tell different people all they need to know about individual plants, entire crops in a particular area, and future yield. We can even improve the yield for different parts of one farm. But our images also provide many other insights, such as giving growers an optimal window for harvest; that is, when the plant is going to yield the most nutritious — or most flavorful — produce. 

Picking a crop at its peak can be a matter of a three-day window, which if missed even by a day or two can lead to a substantial drop in yield. Our plant data can also inform the breeding of seeds so that scientists can dial in acreage production to an even greater degree.

For what kinds of crops do people currently use your imaging technology to improve yields?

Specialty crops — grapes and other fruits — are where we play most. By far our largest group of customers is vineyards. Adopting technology early can be expensive, but grapes are a very high-value specialty crop, so vintners could easily earn out the cost by harvesting more grapes at peak flavor. 

Until working with us, they’d largely relied on anecdotal evidence of vine health gathered by growers examining plants by hand. Now, our FLASH technology scans entire vineyards twice a day, which produces as much as two terabytes of data per camera. 

Connectivity can be hard to get out in the field, so one of our biggest challenges quickly became the transfer of data. Some customers have to express -mail their data to us on disks. When working on optimizing harvests, and the window is small, this data issue becomes an even more urgent problem to solve.

Is that something that winning Akamai’s challenge can address?

Definitely. In addition to some other things we are doing — better compression of the raw images and more preprocessing on the camera — Akamai’s edge compute and real-time data processing would open substantial new markets and opportunities for us. 

Security is the other huge issue that Akamai will help us solve. What we are selling is data and knowledge, both of which will be used by an array of entities, from farmers to scientists to possibly even regulatory authorities. Keeping the intellectual property of our various customers separate, and keeping our own data secure during transfer — not to mention secure from attacks — is paramount. 

The other part is the flexibility that Akamai’s range of solutions offers us. We rely on a hyperscaler machine learning platform tech for back-end tasks, but we are keen to see if we might use Linode for the more portable processing elements. We don’t want to be reliant on one platform, and Akamai’s containerization allows for amazing flexibility. 

Finally, let’s talk about the future of Bloomfield. You’ve won a contract with NASA to help them grow food on the International Space Station. How exciting is that?

We’re thrilled. I mean, to be real, it’s in the proof-of-concept stage. The aim is to feed travelers to the moon and Mars, but right now it’s happening on Earth in a lab. That’s always the first step. But there are so many benefits to being able to grow more crops in space, more varieties, with better yields. 

Fresh crops have many benefits for astronauts: oxygen production, for one, plus the high nutrient content of fresh food, which benefits not only their bodies but their psychological health. These benefits are what we are looking to achieve here on Earth, too. 

We not only need to give more people consistent access to food but access to healthy, nutritious, fresh food. That’s our aim: to feed the world better. We’re making progress, and winning this challenge is a huge leg up to getting there.

Find out more 

Learn more about the $1M Future of Life Online Challenge.