Detailansicht des Stellenangebots

Beschreibung:

Type of activity: Pioneering effort to create a new programming paradigm for parametric and implicit geometry that will power an AI foundation model for turbomachinery. You will architect geometry generation frameworks, build scalable datasets, and connect them to multi-physics simulation chains. Initially focused on turbine blade internal cooling features, your work will evolve toward handling large geometry models and robust automation across CAD/CAE and HPC environments.
How You’ll Make an Impact
• Design and implement a parametric and implicit geometry programming framework for turbomachinery components (e.g., turbine blades, vanes), enabling systematic feature variation and high-fidelity shape control.
• Build and curate large-scale geometry datasets (including metadata, schemas, and versioning) to train and evaluate an AI foundation model for geometry; ensure data quality, consistency, and IP compliance.
• Develop methods to represent geometry for AI (e.g., CAD kernels, constructive solid geometry, signed-distance/field-based models, meshes/point clouds) and evaluate model performance and generalization.
• Integrate and automate multi-physics simulation chains (CFD/CHT/FEA) to generate physics-based labels and performance metrics for training and validation, including robust meshing and model setup.
• Create surrogate and reduced-order models for thermal and structural performance; link them to design optimization workflows and digital twin concepts.
• Implement scalable submission of large simulation batches to HPC/grid environments; optimize pipelines for reliability, throughput, and cost (schedulers, containers, job arrays, CI/CD).
• Collaborate with CAD/CAE, cooling design, and AI teams to capture requirements, validate models, and ensure usability and accuracy of tools in production design environments.
• Establish documentation, coding standards, and training materials; mentor users and drive adoption across engineering teams.

Requirements
• PhD/MSc in Mechanical Engineering, Aerospace Engineering, Computer Science, or a closely related field, with emphasis on geometry processing, CAD/CAE, or thermo-fluids. On top of PhD/Promotion no further professional experience is required;
• Solid understanding of internal cooling for turbine blades (e.g., serpentine channels, impingement, pin-fin arrays, film cooling) and the basics of CHT in turbomachinery.
• Strong programming skills: Python and C/C++; proficiency with Linux and GitLab; experience with API-driven CAD/CAE automation (e.g., NX Open, Star-CCM+ macros).
• Experience with AI/ML applied to geometry and/or physics (e.g., mesh/graph neural networks, diffusion models, surrogate modeling), including dataset design, training, and validation at scale.
• Familiarity with CFD, CHT, and FEA workflows; ability to set up, automate, and validate simulations for thermal and structural performance.

• Mobiles Arbeiten nach individueller Absprache möglich zu ca. 40% mögl.