In this position, you will work in an interdisciplinary team of HPC, ML, and biomedical experts. You will contribute your individual strengths and develop further in related fields. Depending on your profile, you can set priorities—e.g., scalable training, data/feature pipelines, or model-based development—and work closely with Helmholtz Munich, MDC Berlin, and NVIDIA.
You will
- Set up and optimize large-scale training and inference runs for Foundation Models on JUPITER (multi-GPU/-node, mixed precision, parallelization, I/O optimization)
- integrate multimodal data sources (e.g. scRNA-seq, ATAC-seq, spatial transcriptomics, etc.) into the modeling and develop powerful data/training pipelines
- Advancing use cases with Helmholtz Munich, MDC Berlin and NVIDIA – such as disease modeling, prediction of cell states or patient stratification
- contribute to the further development of model architectures (e.g. transformers, multimodal fusion, generative models)
- Present results at conferences and publish in peer-reviewed journals
Table of Content
Summary
Subscribe for Scholarship Alert!
Benefits
- KNOWLEDGE & TRAINING: Comprehensive training offers and individual opportunities for personal and professional development
- HEALTH & WELL-BEING: Your health is important to us. Look forward to a comprehensive company health management program with a wide range of offerings—e.g., a beach volleyball court, running groups, yoga classes, and much more. In addition, our company medical service and an experienced social counseling team are available on-site to assist you.
- WORK-LIFE-BALANCE: The possibility of flexible working (in terms of location) is generally available after consultation and in accordance with the upcoming tasks and (on-site) appointments
- FLEXIBILITY: Flexible working time models, including near-full-time options, allow you to customize your working hours. Learn more at: https://go.fzj.de/vollzeitnah
- HOLIDAY: You will receive 30 days of holiday plus free bridge days (e.g. between Christmas and New Year)
- FAIR COMPENSATION: Depending on your existing qualifications and the tasks assigned, you will be placed in pay group 13 of the TVöD-Bund (Federal Public Service Agreement).
Requirements
- A master's degree, preferably followed by a PhD in computer science, natural sciences, mathematics, or similar, with excellent academic qualifications, documented by publications in relevant journals; a strong quantitative/statistical background (e.g., probability theory, linear algebra, optimization)
- Strong expertise in at least one of the following areas and willingness to deepen the others within the team: High-performance computing (distributed systems, profiling, performance optimization), training of large AI models (PyTorch/JAX/TensorFlow, parallelization, mixed precision), data analysis/statistics (experiment design, evaluation, uncertainty quantification)
- Very good programming skills in Python (C/C++ is an advantage); experience with HPC environments (e.g., SLURM) is welcome
- Interest in biological questions and enjoyment of interdisciplinary collaboration
- Very good English skills; knowledge of German is a plus
- Experience with multimodal omics data (e.g. scRNA-seq, ATAC-seq, spatial transcriptomics, CITE-seq) or medical imaging data (H&E) would be desirable.
- Contributions to open source projects, benchmarks or repro pipelines would also be desirable
Application Deadline
Not SpecifiedHow To Apply
Are you qualified and interested in this opportunity? Kindly go to
Forschungszentrum Jülich on www.fz-juelich.de to apply
For more information, kindly visit FZ Julich scholarship webpage.
