A Simulation Engineer working in AI and machine learning brings two disciplines together. They understand the physics behind computational fluid dynamics, structural analysis and multiphysics simulation, and they know how to use machine learning to make that simulation faster, cheaper and easier to use in design decisions. Depending on the team, the same role might be advertised as CFD Machine Learning Engineer, CAE Machine Learning Engineer, FEA Engineer (Machine Learning), Multiphysics Simulation Engineer, Computational Engineer, Engineering Simulation Scientist or Digital Twin Engineer.
Companies hire for this role when traditional simulation becomes a bottleneck. A single high-fidelity CFD or crash run can take hours or days on a cluster, which limits how many design options an engineering team can explore. ML-based surrogate models trained on simulation and test data can return useful predictions in seconds, so engineers can test far more ideas before committing to expensive runs or physical prototypes.
What does a Simulation Engineer (AI/ML) do?
The work sits between the simulation team, the data science team and the engineers who make design decisions. A Simulation Engineer builds the data pipelines and models that connect them, and makes sure the results are trustworthy enough to act on.
- Setting up and running high-fidelity simulations in CFD, FEA and multiphysics solvers to generate training data
- Building surrogate and reduced-order models that approximate solver outputs at a fraction of the cost
- Applying physics-informed neural networks, neural operators and graph neural networks to meshes and fields
- Designing sampling strategies and active learning loops so expensive simulations are spent where they add most information
- Validating ML predictions against solver results and test data, and quantifying uncertainty before results are used in design
- Building digital twins that combine sensor data with simulation models of a product or asset in service
- Packaging models into tools and workflows that design engineers can use without being ML specialists
Key skills and backgrounds
Strong candidates usually start from one side and grow into the other. Some are CFD or FEA engineers with a mechanical or aerospace engineering degree who have taught themselves machine learning. Others are ML engineers or data scientists who have spent years working with physical systems. Many hold a PhD in computational mechanics, fluid dynamics or applied mathematics. The best ones can explain both why a turbulence model or mesh choice affects the data and why a model architecture suits the problem.
When interviewing, ask how a candidate checked that a surrogate model was safe to use outside the data it was trained on. The answer shows whether they think like a simulation engineer as well as a data scientist.
- Solid grounding in fluid dynamics, solid mechanics, heat transfer or multiphysics, depending on the application
- Hands-on experience with solvers like OpenFOAM, Ansys Fluent, STAR-CCM+, Abaqus, LS-DYNA or COMSOL
- Python with PyTorch or JAX, plus scientific computing libraries
- Surrogate modelling, reduced-order modelling and uncertainty quantification
- Physics-informed ML methods, neural operators and graph neural networks for mesh-based data
- Meshing, pre and post-processing, and automating simulation workflows on HPC or cloud infrastructure
- Clear communication with design engineers who need to know how far to trust a prediction
Where Simulation Engineers (AI/ML) work
The role appears wherever products are designed through simulation. Automotive and motorsport teams use it for aerodynamics, thermal management and crash performance. Aerospace and defence organisations apply it to airframes, propulsion and structures. Energy companies hire for wind, nuclear, oil and gas and battery applications. Semiconductor, industrial equipment and consumer product businesses use it for thermal and structural design. Simulation software vendors and engineering consultancies also build dedicated ML simulation teams.
Simulation Engineer (AI/ML) vs related roles
Several roles overlap with this one, and the title often depends on the team rather than the work.
- A traditional CFD or FEA Engineer runs and interprets simulations directly, while a Simulation Engineer (AI/ML) also builds the ML models that learn from those simulations.
- A Scientific Machine Learning Researcher develops new methods, like novel neural operator architectures, where this role applies and adapts proven methods to engineering problems.
- A Machine Learning Engineer builds and deploys models across many domains, usually without the physics and solver expertise this role depends on.
Salary and market data
Pay depends on seniority, the depth of simulation expertise and the sector, with automotive, aerospace, energy and software vendors each benchmarking differently. Our AI and deep tech salary guides cover adjacent ML and research roles, and the deep tech salary calculator helps you check an offer.
Hiring Simulation Engineers with DeepRec
DeepRec recruits for AI4Science and machine learning teams across Europe and the US, from offices in London, Berlin, Zurich, Dublin and Boston. People who are genuinely strong in both simulation and machine learning are hard to find, and many come from adjacent fields, so we screen for both halves of the skill set and look beyond job titles. We hire on a permanent, contract, retained and executive search basis. To start a search, visit Looking to Hire.