WE ARE DELIVERING

Deep living maps of the planet

TellusLabs combines decades of satellite imagery with a machine learning platform to answer critical, time-sensitive economic and environmental questions.

Timely and accurate agricultural intelligence

In 2016, TellusLabs’ Kernel Beta product predicted end-of-season US Soy yield with 0% error (and came within 1% of US Corn).

On July 1, we are releasing version two. This Kernel is an immersive software product that presents the output of our best-in-class crop forecasting models alongside contextual remote sensing, weather and ground information.

Our first cohort of customers can access Kernel via API and web application.

Why this matters now

Planetary-scale problems demand planetary-scale solutions

Getting insights from satellite data is harder than it should be. Critical emerging challenges like food security, water scarcity, and sustainable urban and industrial systems depend on a view from above. The challenge is that these key Earth data assets are stranded. They are not analytics-ready, critical questions go unasked, and the answers that do arrive are not sufficient to drive meaningful change.

We are changing this with our planetary-scale data analytics platform

How it works

From imagery to insight

Satellite Observations

Data collection

Today humans are more than four decades into the era of spaceborne Earth Observation. Each day, dozens of government and commercial spacecraft add to this record, each of these instruments capturing a unique slice of the always-changing planetary environment.

EARTH OBSERVATIONS DATABASE

A staggering quantity of information

The best models require an integrated, constantly-refreshed global mosaic that spans many instruments, geographies, and timescales. Even five years ago, the sheer compute and storage required meant only large corporations and governments could do this. Distributed databases and cloud compute have changed that.

EARTH OBSERVATIONS DATABASE

A staggering quantity of information

The best Earth models require an integrated, constantly-refreshed global mosaic that spans many instruments, geographies, and timescales. Even five years ago, the sheer compute and storage required meant only large corporations and governments could do this. Distributed databases and cloud compute have changed that.

TELLUSLABS’ EARTH FEATURES DATABASE

Strong features make for strong models

Experienced data scientists focus a lot on feature engineering – translating raw data into meaningful representations of reality that are problem-specific. We have learned the hard way what works (and what doesn’t) for translating raw Earth data into real vital signs for the planet. TellusLabs’ features database is a big part of our story.

Forecasts and Insights

Delivering answers that matter, everyday

Above all, we are an applied sciences company, which means a strong bias towards practical, results driven science and engineering. Our multidisciplinary team combines experience in remote sensing, meteorology, and plant bio-physiology. Our belief is that world-class forecasts and insights come from deep human expertise as well as machine learning. Having the right answer is not always enough; we are equally committed to building insights platforms that help our customers make better decisions.

Forecasts and Insights

Delivering answers that matter, everyday

Above all, we are an applied sciences company, which means a strong bias towards practical, results driven science and engineering. Our multidisciplinary team combines experience in remote sensing, meteorology, and plant bio-physiology. Our belief is that world-class forecasts and insights come from deep human expertise as well as machine learning. Having the right answer is not always enough; we are equally committed to building insights platforms that help our customers make better decisions.

As covered in

National news coverage and awards