Senior Machine Learning Software Engineer in Applied Physics

Join our dynamic team at nTop, where we’re revolutionizing the engineering landscape with our innovative technology. We’re seeking a Senior Machine Learning Software Engineer in Applied Physics to contribute to cutting‑edge projects that greatly enhance the efficiency of engineering design and simulations.

In this role, you’ll tackle complex challenges and work collaboratively within a team that prioritizes learning and growth. Your contributions will help us accelerate physics simulations by 100‑1000 times, drastically improving real‑time engineering design and optimization processes.

• Develop and deploy surrogate models to replace intensive physics simulations, achieving impressive speedups of 100‑1000x.
• Integrate current machine learning solutions seamlessly into the nTop technology ecosystem.
• Prepare and curate training datasets by annotating simulation outputs and extracting relevant physics features from multiple fidelity sources.
• Conduct rigorous validation of models against high‑fidelity simulations and experimental data, including uncertainty quantification.
• Document model architectures, assumptions, and limitations clearly for both technical and non‑technical stakeholders.

• MS/PhD in Physics, Applied Mathematics, Mechanical/Aerospace Engineering, or a related computational area.
• 3+ years of experience developing surrogate or reduced‑order models for physical systems.
• Significant expertise in deep learning frameworks (such as PyTorch or Tensor Flow) applied to scientific problems.
• Strong foundational knowledge in numerical methods, PDEs, and computational physics like FEM or CFD.
• Proficient in the Python scientific computing stack:
  Num Py, Sci Py, Pandas, and scikit‑learn.
• Experience with uncertainty quantification and Bayesian techniques in machine learning models.
• Demonstrated capability in handling large‑scale simulation data within high‑performance computing environments.
• Hands‑on experience with at least one physics simulation software package (ANSYS, COMSOL, OpenFOAM, etc.).

• Familiarity with neural operators (such as DeepONet and Fourier Neural Operators) for PDE solving.
• Experience with NVIDIA’s ecosystem for physics AI (including Physics Nemo, Omniverse, Sim Net).
• Knowledge of NVIDIA Domino for machine learning model management and deployment.
• Experience in setting up and deploying inference services.
• Proficiency in containerization technologies, especially Docker.
• A background in aerospace and energy systems is a plus.
• GPU programming experience (using CUDA or OpenCL).
• Understanding of digital twin architectures and requirements for real‑time simulation.

• €70.000‑€80.000 annually plus options.

nTop

Senior (5+ years of experience)

• Industry
• Machine Learning
• NLP
• United Kingdom