ML Application Expert

VN 24 / 55 AI / ML Application Expert

EUMETSAT uses data from its growing earth observation satellite fleet and from other agencies to support weather forecasting (nowcasting and numerical weather prediction) of weather 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 series of satellite observations.

EUMETSAT is supporting its users utilising AI / ML directly and by developing data products for the applications such as characterisations of weather and climate-related features, including convective storms, flash floods, and fog, in close collaboration with the users.

Within the Climate Services Team of the User Support and Climate Services Division (USC), the AI / ML Applications Expert exploits satellite radiance data and derived products employing AI / ML methods to deduce weather and climate information, which includes the identification of Earth system features also in combination with other data types.

In addition, the post holder will support users to use EUMETSAT ML-ready data in their applications and establishes collaborations within the community.

The Climate Services team consists of staff with a wide range of expertise from climate applications of Earth Observation data to informatics that produce climate data efficiently.

The team uses the latest technologies that can be applied to cloud / bulk data processing systems to generate and exploit the datasets.

The team also supports the further development of these systems. Joining EUMETSAT in this role offers a unique opportunity to develop a professional career in a highly motivating environment.

Duties

• Systematically explore user needs regarding data content and quality for usage of AI / ML for nowcasting, weather forecasting as well as climate services and science;
• Exploit existing satellite radiance data and derived products employing ML methods to deduce weather and climate information;
• Analyse the performance of existing ML models and perform sensitivity analyses on the content and quality of the input data for the models;
• Maintain knowledge of the best available ML methods for processing EO-data in the context of observation-based weather and climate information, climate modelling, and climate impact predictions;
• Support users to use EUMETSAT data in their applications with guidance, examples and by preparing the data for the ML-applications;
• Collaborate closely with other EUMETSAT teams involved in AI / ML activities;
• Collaborate closely with EUMETSAT member states and the EUMETNET E-AI Programme and explore and establish collaboration within international agency and research settings.

LOCATION

Darmstadt, Germany

QUALIFICATIONS

A master or higher degree in a relevant natural or computer science discipline.

LANGUAGES

The official languages of EUMETSAT are English and French. Candidates must be able to work effectively in English and have some knowledge of French.

DEADLINE

14 January

Duties

• Systematically explore user needs regarding data content and quality for usage of AI / ML for nowcasting, weather forecasting as well as climate services and science;
• Exploit existing satellite radiance data and derived products employing ML methods to deduce weather and climate information;
• Analyse the performance of existing ML models and perform sensitivity analyses on the content and quality of the input data for the models;
• Maintain knowledge of the best available ML methods for processing EO-data in the context of observation-based weather and climate information, climate modelling, and climate impact predictions;
• Support users to use EUMETSAT data in their applications with guidance, examples and by preparing the data for the ML-applications;
• Collaborate closely with other EUMETSAT teams involved in AI / ML activities;
• Collaborate closely with EUMETSAT member states and the EUMETNET E-AI Programme and explore and establish collaboration within international agency and research settings.

Skills and Experience

• Proven experience with scientific satellite applications involving large scale data analytics;
• Both theoretical and practical understanding of data-driven methods and their validation;
• Proven capability to handle complex problems and tasks independently with can-do attitude;
• Experience with at least one of the relevant ML libraries (Scikit Learn, TensorFlow, Keras, or PyTorch);
• Good programming skills in at least Python;
• Excellent interpersonal and communications skills and willingness to work in direct contact with users, combined with a proven ability to work within a team;
• Experience in working in the international weather and climate domain is seen as an advantage.

Employment Conditions

The initial contract will be of 4 years’ duration, with subsequent 5 year contracts being awarded thereafter, subject to individual performance and organisation requirements.

There is no limit to the amount of follow-up contracts a staff member can receive up to the EUMETSAT retirement age of 63 and there are certainly opportunities to establish a long career perspective at EUMETSAT.

This post is graded A2 / A4 on the EUMETSAT salary scales. The minimum basic salary for this post is EURO per month (net of internal tax but excluding pension contribution and insurances) which may be negotiable on the basis of skills and experience.

The salary scale provides for increments on the anniversary of taking up employment, and scales are reviewed by the EUMETSAT Council with effect from 1 January each year.

In addition to basic salary, EUMETSAT offers attractive benefits. Further information, including salary details, is available on the EUMETSAT web site.

EUMETSAT is committed to providing an equal opportunities work environment for men and women.

Please note that only nationals of EUMETSAT Member States may apply. The EUMETSAT Convention requires that Staff shall be recruited on the basis of their qualifications, account being taken of the international character of EUMETSAT.