The DLR Institute of Atmospheric Physics studies the physics and chemistry of the global atmosphere from the ground to the top of the middle atmosphere at an altitude of about 120 km. The department Earth System Model Evaluation and Analysis develops innovative methods for the evaluation and analysis of Earth system models in comparison with observational data with the goal of better understanding and predicting the Earth system. In cooperation with international partners, the institute is leading the development of a software tool to check the quality of complex Earth system models using observational data (ESMValTool). As freely available software, the ESMValTool supports model evaluation and the achievement of a better process understanding within the Coupled Model Intercomparison Project (CMIP). Together with partners from Jena, Valencia and New York, the institute has won the prestigious European Research Council (ERC) Synergy Grant for the project "Understanding and Modelling the Earth System with Machine Learning (USMILE)". Within this project, we aim to close gaps in our understanding of the climate system in the area of small-scale physical processes, such as clouds. The goal is to better understand and model changes in the climate system and their effects on the Earth's ecosystems using machine learning methods. For this purpose, we are looking for a research software engineer to become part of the international development team for the ESMValTool (core developer) and USMILE.

Your tasks will include the following:

• Technical execution of simulations using the ICON-ML model and technical support for related developments.
  • Technical support for the development of ICON-ML
  • Technical execution of simulations with ICON-ML on various large-scale computing systems
  • Technical support of department members in the development of ICON-ML, use of ICON-ML model and simulations
• Technical adaptation and further development of machine learning techniques for the evaluation of Earth system data and the improvement of the ICON-ML model
  • Provide technical support to department members in developing innovative methods for analyzing large data sets, including incorporating machine learning techniques into the ESMValTool and testing them
  • Technical support of the department members in the implementation and development of machine learning methods for ICON-ML and on different mainframe systems (CPU/GPU)
• Technical further development of the ESMValTool with focus on optimization of the software
  • Extension and optimization of the technical infrastructure of the ESMValTool
  • Porting to computers connected to Earth System Grid Federation (ESGF)
  • Support of the ESMValTool development cycles, software engineering and bug analysis
  • Development and execution of regular tests on different computer systems
  • Development of efficient evaluation procedures with respect to increasing data volume (petabyte requirements)
  • Technical support of ESMValTool application developers (also international)