Dr. Susann Hänsch Tel.: +49 351 260 3192

Job-Id: 2025/93 (2092)

At HZDR, we promote and value diversity among our employees. We welcome applications from people with diverse backgrounds regardless of gender, ethnic and social origin, belief, disability, age, and sexual orientation. Severely disabled persons are given preference in the event of equal suitability.

Helmholtz-Zentrum

Postdoc (f/m/d) AI-based consulting system for applied multiphase CFD simulations

With cutting-edge research in the fields of ENERGY, HEALTH and MATTER, around 1,500 employees from more than 70 nations at Helmholtz-Zentrum Dresden-Rossendorf (HZDR) are committed to mastering the great challenges facing society today.

The Institute of Fluid Dynamics is conducting basic and applied research in the fields of thermo-fluid dynamics and magnetohydrodynamics in order to improve the sustainability, the energy efficiency and the safety of industrial processes.

The Department of Computational Fluid Dynamics is looking for a Postdoc (f/m/d) AI-based consulting system for applied multiphase CFD simulation.

We have developed a comprehensive database of computational fluid dynamics (CFD) simulation cases and are currently creating a performance matrix to evaluate CFD closure models. The over-arching goal is to conserve the experience gained with every CFD simulation. This data-driven approach allows us to apply machine-learning techniques to infer connections between the feature space of our CFD cases, closure models and the performance of their interaction, based on which a recommender system is developed to predict the best model set for new CFD cases.

Your tasks

• Identify, implement, test and ensemble suitable recommender algorithms (collaborative/content-based filtering etc.)
• Strategy to collect and store relevant data for the long-term built-up of a performance database (for the interaction between CFD case and CFD closure model)
• Development of a performance metric characterizing the speed of each model
  combination (computed from the runtime statistics on our HPC cluster)
• Use multiple matrices (i.e. for accuracy, robustness and speed performance) for different
  user needs to allow for a flexible usage of the recommender system application
• Development of a strategy for user feedback with given recommendations (explicit/implicit)

Your profile

• Completed university studies (Master or PhD) in the field of data sciences or related field
• Strong foundational knowledge about recommender systems and associated algorithms
• Experience with data retrieval and storage to build a sustainable dataset
• Out-of-the-box attitude to newly apply machine-learning methods to a natural science field
• Excellent programming knowledge in Python
• Excellent language skills (written + verbal English)

Our offer

• A vibrant research community in an open, diverse and international work environment
• Scientific excellence and extensive professional networking opportunities
• Salary and social benefits in accordance with the collective agreement for the public sector ( TVöD-B und) including 30 days of paid holiday leave, company pension scheme (VBL)
• We support a good work-life balance with the possibility of part-time employment, mobile working and flexible working hours
• Numerous company health management offerings
• Employee discounts with well-known providers via the platform Corporate Benefits
• An employer subsidy for the "Deutschland-Ticket Jobticket"

We look forward to receiving your application documents (including cover letter, CV, diplomas/transcripts, etc.), which you can submit via our online-application-system.