Öffnet in neuem Fenster Opens in a new window Öffnet externe Seite Opens an external site Öffnet externe Seite in neuem Fenster Opens an external site in a new window

Department Optics and Beamlines

Scientific IT Consulting

In a unique collaboration, Berlin University Alliance (BUA), the Max Delbrück Center (MDC), and the Helmholtz Centre for Materials and Energy Berlin (HZB), together with the Zuse Institute Berlin (ZIB), aim to establish a powerful Data and AI Center in the German capital. The Joint Berlin Data & AI Center planned at the Adlershof campus will house the computational, storage, and networking resources necessary for next-generation experiments, particularly those involving the upcoming BESSY III synchrotron. By integrating this cutting-edge hardware platform with ZIB’s existing HPC capabilities, the high-speed BRAIN network (up to 400 Gbit/s), and extensive experience in scientific software engineering, we provide HZB researchers with a seamless, end-to-end research environment. This spans from raw detector data to AI-enhanced insights, all managed under strict adherence to FAIR (Findable, Accessible, Interoperable, Reusable) data principles.

At Scientific IT Consulting, we serve as the collaborative hub between the Helmholtz Center Berlin and the Zuse Institute Berlin. Our mission is to translate HZB’s scientific vision into practical, reproducible data-processing solutions that leverage ZIB’s proven expertise in high-performance computing (HPC) and artificial intelligence (AI) infrastructure.

Our specific objectives include:

  • Development and Application of AI-Based Research Tools: Designing innovative solutions tailored to scientific research needs.
  • Consulting and Support for Researchers and IT Experts at HZB: Guiding researchers in leveraging AI methods across all relevant disciplines, with a focus on those working with synchrotron beamlines and instruments at BESSY II.
  • Establishment of Best Practices: Creating methodologies for AI, HPC, and sustainable reuse of research data throughout the entire Research Data Life Cycle.
  • Digital Twin Development: Creating realistic models for digital twins of experimental methods, including synchrotron beamlines and instruments at BESSY.
  • Real-Time Data Analysis Software: Developing tools for rapid analysis of operational and measurement data to enable automated facility optimization and autonomous experimentation.
  • Efficient Algorithm Development: Designing algorithms that allow for faster and more accurate evaluation of large datasets with reduced resource consumption.

Current status:

In our current concept phase, we are conducting a detailed needs analysis focused on research data handling. This includes efficient data generation, processing, condensation, analysis, storage, exchange, and reuse. In line with Good Scientific Practice (GWP), the research data we manage will be scalable, transparent, and prepared for use in training AI models such as neural networks and other machine learning methods. We are also developing a concept to implement FAIR (Findable, Accessible, Interoperable, Reusable) principles for large datasets—ranging from terabytes to petabytes per year.

Safety and sustainability:

  • Kubernetes (for Docker orchestration),
  • JupyterHub (for code documentation),
  • Zenodo (as a data repository),
  • eLabFTW (for electronic lab notebooks),
  • Bluesky (for Python-based process automation),
  • ImageJ (for image analysis),
  • and other tools

Based on the specific needs of HZB and its partners, we will staff the group with scientific personnel and acquire and document the required software.

The group, led by Dr. Yannic Kerkhoff, was established in January 2026 and is still in development.

© 01/26 Yannic Kerkhoff