Focused
Machine Learning Engineer (f/m/d) at Focused
Role overview
An at-a-glance view of the facts included with this Machine Learning Engineer (f/m/d) at Focused listing.
- Company
- Focused
- Location
- Darmstadt, DE
- Work arrangement
- On-site
- Employment type
- Full-time
Applications for this listing are submitted through CareersForge. Use the form on this page, then sign in to your Candidate Portal to follow application updates and any interview invitation.
Role overview
Focused Energy is developing laser-driven inertial fusion systems, and this Machine Learning Engineer will work across its Software Engineering and Digital Twin environments. The role combines machine learning engineering with computational physics, partnering with optical engineers, laser and fusion physicists, simulation scientists, and systems engineers to incorporate data-driven methods into the Digital Twin architecture.
You will build and deploy models that augment or accelerate multiphysics simulations for high-energy laser systems, target injection and tracking, target manufacturing, and the fusion chamber. The work supports faster design cycles, system optimization, and decision-making using simulation results and sensor data.
The scope includes surrogate and reduced-order models, physics-informed machine learning, scientific ML pipelines, active learning, Bayesian optimization, anomaly detection, and approaches to autonomous or semi-autonomous laser parameter optimization. You will also integrate models with HPC simulation outputs—including COMSOL, ANSYS, and custom solvers—and help establish practices for model testing, versioning, reproducibility, and documentation. The position is listed as full-time and on-site in Darmstadt, Germany.
Role at a glance: Full-time, on-site Machine Learning Engineer role in Darmstadt focused on ML for multiphysics simulations and fusion-energy Digital Twin systems.
Responsibilities mentioned in the listing
- Design and deploy surrogate and reduced-order models to replace or speed up high-fidelity multiphysics simulations.
- Develop physics-informed machine learning models, including models that encode physical constraints such as Maxwell’s equations, thermodynamics, and fluid dynamics.
- Build and maintain ML pipelines for training, validation, uncertainty quantification, and ongoing refinement using experimental and simulation data.
- Create active-learning and Bayesian-optimization workflows to guide design exploration and reduce simulation runs.
- Integrate ML models into the Digital Twin framework, connecting them with HPC simulation outputs, including COMSOL, ANSYS, and custom solvers, as well as real-time sensor data.
- Develop anomaly-detection and predictive-diagnostics models for laser-system health and off-nominal behavior.
- Apply reinforcement learning and Bayesian control methods to support autonomous or semi-autonomous optimization of laser operating parameters.
Qualifications stated in the listing
- Master’s or PhD in machine learning, computational physics, applied mathematics, data science, computer science, or a closely related field.
- Demonstrated experience building, training, and deploying ML models for complex physical systems; strong knowledge of deep learning frameworks such as PyTorch, TensorFlow, or JAX, probabilistic models, and uncertainty quantification.
- Expert-level Python skills; C++ or Fortran proficiency is a plus.
- Experience with HPC environments, including batch schedulers and MPI or OpenMP parallelization.
- Ability to work with PDE-based simulation outputs, time-series sensor data, and high-dimensional parameter spaces.
- Hands-on experience with Gaussian processes, neural-network surrogates, reduced-order models, or comparable metamodeling methods.
- Experience creating scientific-data ML pipelines, including preprocessing, feature engineering, validation, and deployment.
- Ability to explain model behavior, confidence intervals, and limitations to physics, engineering, and non-ML audiences.
