Machine Learning Application Developer

EUMETSAT; Darmstadt; Vor Ort; EUR 67.000 - 89.000; Vollzeit

Benefits dieser Stelle

Relocation support

Training and development

Flexible working time

Medical coverage for employee and fam

Annual leave + holidays

Zusammenfassung

EUMETSAT seeks an experienced Machine Learning Application Developer to exploit satellite data using AI/ML methods, focusing on weather and climate feature detection. You will build data pipelines, train models, and validate results to support forecasts and climate services.

The role involves collaboration across an international team, working with EO data, ML pipelines, and modern software practices. English proficiency and willingness to relocate to Darmstadt, Germany, are essential.

Qualifikationen

• Master’s or higher degree in a relevant computing or natural sciences subject.
• Experience in ML model development and evaluation, preferably in weather and climate or Earth observation domain.
• Experience with at least one ML library (Scikit Learn, TensorFlow, Keras, PyTorch).
• Strong programming skills in Python and some experience with compiled languages; Git-based development workflows, CI/CD, code review, and automated testing.
• Experience with large datasets.
• Experience in working in cloud environments.
• Excellent communication skills, with the ability to work effectively in a team.
• Candidates must be able to work effectively in English.

Aufgaben

• Support internal and external users by creating data pipelines and curated ML-ready datasets for each use case.
• Develop, train, apply, and validate ML models to derive weather/climate information.
• Explore and share methods to access EO data and apply models.
• Encourage sharing of code samples, datasets, and pre-trained models within the community.
• Support training courses, hackathons, and other community activities.

Kenntnisse

ML model development

ML libraries

Python programming

Large datasets

Cloud computing

Team collaboration

Communication skills

Ausbildung

Master’s degree

Tools

MLFlow

Kubeflow

Git

Jobbeschreibung

EUMETSAT is Europe’s meteorological satellite agency - monitoring the weather and climate from space - 24 hours a day, 365 days a year.

At EUMETSAT, you can make a world of difference and be a part of something that makes a positive impact on society. You will be at the cutting edge of satellite technology, with a meaningful role in an organisation focused on space-based observations of the Earth’s weather and climate.

Background

EUMETSAT generates and uses data from its growing fleet of Earth observation satellites, together with data from partner agencies, to support weather forecasting—including nowcasting and numerical weather prediction—by meteorological and other environmental agencies in its Member States and worldwide. In addition, EUMETSAT supports climate monitoring by providing a wide range of satellite data, including rescued, recalibrated and reprocessed long-term series of satellite observations. EUMETSAT also supports its users in exploiting AI/ML by applying these methods to its core data products and through dedicated data curation activities.

Within the Climate Services Team of the User Support and Climate Services Division (USC), the Machine Learning Application Developer exploits satellite data using AI/ML methods to derive weather and climate information, including the identification and characterisation of Earth system features such as convective storms and fog. In addition, the post holder contributes to the development of the weather forecasting and early warning value chain, as well as to kilometre-scale climate modelling, by supporting users in integrating EUMETSAT ML-ready data into their applications and by establishing collaborations across the wider community.

The team brings together staff with a wide range of world-class expertise, including weather and climate as well as Earth observation domain experts, data processing engineers, and machine learning specialists. Joining EUMETSAT in this role offers a unique opportunity to develop your professional career in a highly motivating international environment.

What you’ll be doing

• Support EUMETSAT internal and external users by creating data pipelines and curated ML-ready datasets with needed data content, quality, format, and structure for each use case;
• Develop, train, apply, and validate machine-learning models and other algorithms to transform Earth observation data into actionable information, particularly for Earth system feature detection;
• Explore and share the best ways to apply modern techniques and access Earth Observation data with the aim of helping the community to develop, train, and apply the models;
• Foster community development by encouraging and assisting EUMETSAT users in sharing commonly applicable parts of their applications in the form of how-to articles, code examples, infrastructure configurations, pre-trained models, datasets, etc;
• Take an active role in user communities and support training courses, hackathons, and other community-oriented means of spreading practical knowledge to EUMETSAT member states.

What we offer

• Excellent salary, of up to Euro 8000 NET (after income tax) based on skills and experience
• Flexible working time including additional flexi-leave
• Full medical coverage for employee and family
• 30 days of annual leave + 14.5 days public holidays
• Training and development support
• Relocation allowance and support (if applicable)

Requirements

Qualifications

Master’s or higher degree in a relevant computing or natural sciences subject.

Skills and Experience

• Experience in ML model development and evaluation, preferably in weather and climate or Earth observation domain;
• Experience with at least one of the relevant ML libraries (Scikit Learn, TensorFlow, Keras, PyTorch…);
• Proven strong programming skills in at least Python and some experience in working with compiled languages and with modern collaborative software development practices, including Git-based development workflows, CI/CD pipelines, code review, and automated testing;
• Experience in working with large datasets;
• Experience in working in cloud environments;
• Eagerness to learn and expand one’s own knowledge, combined with ease in joining teams to solve problems and communicate resolutions;
• Excellent communication skills, with a proven ability to work effectively as part of a team, strong problem-solving skills, and a positive, can-do attitude.

The below skills and experience are seen as an advantage:

• Experience in working with Earth observation and weather and climate data;
• Experience in using machine learning pipelines and frameworks (MLFlow, Kubeflow, etc.);
• Contributions to the open-source software ecosystems.

Languages

Candidates must be able to work effectively in English.

APPLICATIONS MUST BE SUBMITTED IN ENGLISH. WE WILL BE UNABLE TO CONSIDER APPLICATIONS SUBMITTED IN OTHER LANGUAGES.

EUMETSAT is an international organisation employing staff from across all 30 member states and this role is being advertised on multiple international job boards, but please note that this role would be based in Darmstadt, Germany and would require international applicants to relocate if successful.

More about us

EUMETSAT’s role is to establish and operate meteorological satellites to monitor the weather and climate from space - 24 hours a day, 365 days a year. This information is supplied to the National Meteorological Services of the organisation's Member and Cooperating States in Europe, as well as other users worldwide.

EUMETSAT also operates several Copernicus missions on behalf of the European Union and provide data services to the Copernicus marine and atmospheric services and their users.

As an intergovernmental European Organisation, EUMETSAT can recruit nationals only from the 30 Member States (Austria, Belgium, Bulgaria, Croatia, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, The Netherlands, Norway, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey and the United Kingdom).