Passionate about Climate, Energy and AI? Join Kayrros, as a Junior Remote Sensing Scientist M/F

Kayrros is a global climate technology company and the world leader in environmental intelligence.

Founded in 2016, it uses artificial intelligence and machine learning to analyze massive amounts of satellite imagery and other inputs at scale and generate standardized, near real time data on greenhouse gas emissions and energy supply chains used for trading, environmental compliance and reporting purposes, as well as bespoke data on emissions, climate risks, land-use change and other metrics used by public and private-sector actors such as asset managers, regulators, industrial and consumer-product companies, insurers and others in various proprietary applications. Kayrros technologies provide a powerful set of high-impact tools to reduce our climate footprint, manage climate risks, harden our resilience to extreme weather events, and accelerate the transition to a lower-carbon economy.

Kayrros’s accomplishments have earned it wide recognition. We have been named one of TIME's 100 Most Influential Companies, been featured on Fortune’s ‘Change the World’ list in both 2023 and 2024, named Top Five Company of the Year in Fast Company’s ‘World Changing Ideas’ 2024 awards, and won the Financial Times Tech Champion award.

We have offices in Paris, Houston, New York, London, Bangalore, and Singapore. Now, we’re looking to grow our outstanding team by hiring gifted individuals who want to disrupt global climate governance and shape the future of finance and the environment.

Join us to make a difference!

For more information, visit www.kayrros.com.

We are currently recruiting for our Paris office, an Intern: Junior Remote Sensing Scientist H/F

Project: Active Fire Detection using Geostationary Satellites

Wildfires are becoming more severe, larger in scale, and start earlier than the typical fire season. Early detection and continuous monitoring are more important than ever. In this context, Kayrros is developing tools for real-time monitoring of wildfires using remote sensing data. To complement existing solutions using low-Earth orbit satellites (such as the MODIS, VIIRS, and Sentinel-3), we are working on adding the latest European geostationary satellite to complete and enhance Kayrros’ active fire monitoring service.

Eumetsat’s Meteosat Third Generation (MTG) satellite, which is just starting to become operational, acquires images every 2.5 minutes over continental Europe at a resolution of 4 km/pixel. Although the spatial resolution is coarser, the acquisition rate (or temporal resolution) is significantly higher than the typical 12 hours from low-Earth orbiting satellites. In addition, images are available to users with a very low latency: within a few minutes of acquisition.

The goal of this internship is to develop active fire detection algorithms using the dense time series provided by geostationary satellites such as MTG. Indeed, exploiting the temporal aspect has the potential to greatly aid the detection of small fires, thus increasing the number of fires detected compared to most methods that work on a per-image basis[1]. The approach we are considering builds on Kayrros’ expertise in methane detection in time series, including from geostationary satellites [Groshenry et al. 2024], using change detection and background subtraction, and hotspot detection using anomaly detection [Tadros et al. 2024, Hessel et al. 2024]. This may evolve depending on the preliminary review of the state of the art in the field. The application of the developed algorithm to other geostationary satellites, such as the GOES or Himawari satellites (from the U.S. and Japanese space agencies, respectively), could also be explored to expand the geographic coverage to most of the globe.

In this internship, you will:

• Explore and review the existing approaches and relevant associated scientific literature on active fire detection, in particular from Geostationary satellites;
• Design, implement, and test algorithms;
• Validate the results against other algorithms for the same source, as well as with other sources, for example, the MODIS, VIIRS, and Sentinel-3 satellites;
• Apply these algorithms for real-world utilization in the Kayrros service for active fire monitoring with the help of data engineers.
• Formulate problems considering available data and project constraints.
• Continuously provide ideas to improve products and take part in all team discussions.
• Use your analysis skills to understand the product and project you’ll be working on
• Grow a culture around remote sensing data.