PhD in Geospatial Science, Remote Sensing, Data Science, Computer Science, or a related field
Strong proficiency in Python and geospatial libraries such as GDAL, Rasterio, GeoPandas, and xarray
Experience with remote sensing techniques and platforms/sensors (e.g., Sentinel, Landsat)
Knowledge of machine learning frameworks (TensorFlow, PyTorch, Scikit-learn) and statistical modeling
Requirements:
Conduct a state-of-the-art review of data fusion methods applied to satellite/UAV/IoT data
Develop and test multi-sensor fusion methodologies, considering spatial and temporal resolution, geolocation precision, spectral information, and radiometric quality
Leverage cloud infrastructure for data access and perform an objective comparative study between identified approaches, analyzing performance and presenting results
Collaborate with Phenet academic partners (UCL, INRAe) and interact daily with an international, multidisciplinary team
Job description
ABOUT EARTHDAILY
EarthDaily is revolutionizing the way we understand and monitor our planet. Through cutting-edge Earth Observation (EO) technology and geospatial analytics, we provide unparalleled insights for industries ranging from agriculture to mining, insurance, and government intelligence. Our mission is to build the world’s most advanced change detection system to capture, analyze, and interpret global shifts in near real-time.
THE ROLE
EarthDaily is looking for a PostDoc in remote sensing to join its Data Science team and develop innovative methods for processing multi-scale observations.
You will be involved in the development of methods in the framework of the PHENET European project WP3 (https://www.phenet.eu/en).
You will develop a multi-scale image harmonization framework for the analysis of automatically co-registered multi-resolution datasets (Sentinel-2, UAV, EarthDaily, on-field sensors).
The spectral and radiometric adjustment algorithms should be optimized to scale up surface reflectance from different sensors using physical and/or AI methods. These data fusion algorithms will be provided to enable the synergistic exploitation of IoT and EA time series based on Bayesian multi-sensors fusion and interpolation methods.
In PHENET, the European Research Infrastructures (RI) on plant phenotyping (EMPHASIS), ecosystems experimentation (AnaEE), long-term observation (eLTER) and data management and bioinformatics (ELIXIR) join their forces to co-develop new tools and methods for the identification of future-proofed combinations of species, genotypes and management practices in front of the most likely climatic scenarios across Europe.
YOUR MISSIONS
Conduct a state-of-the-art review of data fusion methods applied to satellite/UAV/IoT data
Develop and test fusion methodologies considering pro and cons of each type of observations (spatial and temporal resolution, geolocation precision, spectral information, radiometric quality…)
Leverage cloud infrastructure and data access
Conduct an objective comparative study between identified approaches
Analyze performance and present results
Collaborate with our Phenet academic partners (UCL, INRAe)
Interact daily with an international, multidisciplinary team
EDUCATION, KNOWLEDGE AND SKILLS
PhD in Geospatial Science, Remote Sensing, Data Science, Computer Science, or a related field.
Autonomous and creative
Strong proficiency in Python and geospatial libraries such as GDAL, Rasterio, GeoPandas, and xarray
Experience with remote sensing techniques and platforms / sensors (e.g., Sentinel, Landsat)
Knowledge of machine learning frameworks (TensorFlow, PyTorch, Scikit-learn) and statistical modeling
Familiarity with cloud computing environments (AWS, Google Cloud, or Azure) and big data processing
Strong communication skills and ability to work in an international, cross-functional team.
Fluent in English (oral and written): meetings with partners and internal are mostly in English.
Additionnal "good to have"
Background in agriculture, agronomy, or environmental science.
Knowledge of deep learning techniques for geospatial applications.
Familiarity with Agile development methodologies.
Proficiency in database management for geospatial datasets.
CONDITIONS
Full-time job from May 2026 (12-15 months) Location: Based in Balma (Toulouse, France). Two days of work from home possible. Mutuelle and lunch vouchers.
