Machine Learning Engineer is an open position at Focused Energy located in Darmstadt, Germany. This opportunity suits professionals with advanced academic credentials in machine learning, computational physics, or related disciplines who want to build data-driven intelligence for clean energy systems.
Role Overview
This position integrates machine learning engineering with computational physics within software engineering and digital twin environments. The professional works alongside physicists, simulation scientists, and engineers to embed intelligence into laser and fusion architectures, minimizing design risks and accelerating development.
Key Responsibilities
Build surrogate and reduced-order models to replace or speed up multiphysics simulations.
Develop physics-informed neural networks incorporating physical constraints like thermodynamics and fluid dynamics.
Construct and maintain machine learning pipelines for model training, validation, uncertainty quantification, and continuous refinement.
Apply active learning and Bayesian optimization workflows to guide design space exploration.
Integrate trained models into digital twin frameworks interfacing with high-performance computing simulation outputs and sensor data.
Create anomaly detection and predictive diagnostics models for monitoring laser subsystems.
Apply reinforcement learning and Bayesian control for autonomous optimization of laser operating parameters.
Partner with cross-functional teams to ensure model fidelity, latency, and uncertainty requirements are met.
Establish engineering practices for model versioning, reproducibility, testing, and documentation.
Requirements
Master degree or PhD in Machine Learning, Computational Physics, Applied Mathematics, Data Science, Computer Science, or a related field.
Experience building, training, and deploying models for complex physical systems with deep learning frameworks.
Expert-level Python skills and familiarity with high-performance computing environments.
Background working with partial differential equation simulation outputs, time-series sensor data, and high-dimensional parameter spaces.
Experience with Gaussian processes, neural network surrogates, reduced-order models, or metamodeling techniques.
Ability to construct robust pipelines for scientific data handling and communicate model behaviors clearly.
Nice to Have
Experience with physics-informed neural networks or neural operators applied to physical systems.
Background in laser physics, plasma physics, or high-energy-density science.
Familiarity with digital twin platforms, multidisciplinary design optimization workflows, and reinforcement learning.
Knowledge of Monte Carlo methods, MLOps tooling, and statistical uncertainty quantification frameworks.
Skills and Tools
Python
C++
Fortran
PyTorch
TensorFlow
JAX
High-Performance Computing
MPI and OpenMP
COMSOL
ANSYS
How to Apply
Interested candidates should apply on the official company careers page.
