Prospera Technologies

The problem: disease found too late

A farmer cannot walk every row of a field every day, so a disease or pest is usually found when the damage has already spread. Prospera, founded in 2014, approached the problem through computer vision: cameras installed in the field and on irrigation systems take pictures continuously, and a model identifies from the images signs of disease, pest damage, nutrient deficiency, growth stage and expected yield, at the level of an area within the field rather than the field as a whole.

Computer vision fed into decision systems

The output is not a report to read but input for decision systems: how much to irrigate and where, when to spray and exactly where. This is the core idea of precision agriculture: treating each plot according to what it needs, instead of irrigating and spraying uniformly.

The acquisition by Valmont

The partnership between Prospera and the American company Valmont Industries, one of the world's largest makers of center pivot irrigation systems, began in 2019. On 5 May 2021 Valmont officially announced a binding agreement to acquire Prospera for about $300 million, funded from cash and an existing revolving credit facility. Valmont CEO Stephen Kaniewski described the acquisition as part of a "commitment to make the farm more efficient", and Prospera CEO Daniel Koppel described it as "the fastest way to scale impactful ag-tech solutions". In its results report for 2025, published on 17 February 2026, Valmont announced that Prospera's operations had ended during 2025. So the acquisition does not indicate independent operation or that the business is still running today.

The logic behind the deal: distribution as much as accuracy

The logic behind the acquisition is a broader lesson: a vision algorithm is worth far more when it sits on hardware that is already installed in the field and connected to power and water than as a standalone product that has to convince every farmer to install something new. Distribution, not just accuracy, is sometimes the deciding factor.

The particular difficulty of agricultural computer vision

The main difficulty in agricultural computer vision differs from that in other fields: no two fields are the same. Lighting, crop variety, growth stage, soil type and even the angle of the leaves change from plot to plot and from season to season, and a model trained in one place does not necessarily work in another. That is why collecting data across many fields and over whole seasons is an integral part of the product, not a preliminary step. Another Israeli company in the field, Taranis, bases its platform on images at the resolution of a single leaf and on machine learning.

Continuing Israel's agricultural tradition

Israel has been identified with agricultural innovation for decades, from drip irrigation onward, and the arrival of computer vision in the field continues the same line: not a new crop, but a more precise way to manage what is already growing. The measured result is usually savings in water and pesticides along with more stable yields, not a dramatic jump in quantity.