PhD Position on Physics-Based Machine Learning Modeling for Materials and Process Design Helmholtz-Zentrum Hereon in Germany

The Helmholtz-Zentrum Hereon is a leading international research institution dedicated to addressing global challenges such as climate change, sustainable use of coastal systems, and improving quality of life through resource-compatible innovations. The Institute of Material and Process Design, part of Hereon, focuses on the sustainable and ecological development of advanced materials and manufacturing processes, with special emphasis on the transportation sector and medical technology. The institute tailors materials and designs manufacturing processes to conserve resources, spanning from fundamental modeling to practical applications.

This 4-year PhD position is based in Geesthacht and centers on physics-based machine learning modeling for materials and process design. The project aims to develop machine learning models for clustering, classification, regression, and reinforcement learning tasks, enhancing or replacing established computational engineering and simulation methods such as the finite element method and constitutive materials models. The research will explore relationships along the composition-process-structure-property-performance chain, enabling stability and control of novel manufacturing processes and achieving desired properties in materials science and engineering. Use cases will focus on lightweight structure manufacturing techniques, supporting innovative materials and process design.

Supervision is provided by Prof. Noomane Ben Khalifa (Hereon/Leuphana University Lüneburg) and Dr.-Ing. Frederic Bock (Hereon). The project combines computer simulations and machine learning to tackle complex problems in mechanical engineering and materials research that are not addressable by classical modeling or machine learning alone. Tasks include developing novel machine learning models (supervised, unsupervised, reinforcement learning), data modeling and assimilation with experimental measurements, applying Explainable AI techniques, evaluating system architectures like Retrieval-Augmented Generation (RAG), implementing pipelines in Python (using PyTorch), validating results collaboratively, and publishing in international journals and conferences.

Applicants must have a master's degree in mechanical engineering, materials science, computational engineering, computer science, applied mathematics, physics, or a related field. Strong programming skills in Python, experience with neural networks and Python-ML libraries (e.g., PyTorch), background in computational mechanics and materials science, and high proficiency in English are required.

The position offers a dynamic research environment with around 1,000 employees from over 60 nations, excellent networking opportunities, a well-connected campus, social benefits per the public service collective agreement, remuneration up to pay group 13 (TV EntgO Bund), 6 weeks holiday per year, company holidays, flexible work options, a PhD Buddy Program, family-friendly policies including childcare facilities, employee assistance programs, and corporate benefits. Severely disabled persons and those equaling severely disabled persons will be considered preferentially.

Application deadline is April 26th, 2026. Interested candidates should submit comprehensive application documents (cover letter, CV, transcripts, certificates) indicating reference number 2026/WD 1 via the provided application link. For further details, visit the application portal.