Project Role : Custom Software Engineer

Project Role Description : Develop custom software solutions to design, code, and enhance components across systems or applications. Use modern frameworks and agile practices to deliver scalable, high-performing solutions tailored to specific business needs.

Must have skills : FEA Simulation

Good to have skills : NA

Minimum 5 Year(s) Of Experience Is Required

Educational Qualification : 15 years full time education

Summary:

We are seeking a skilled CFD - Physics AI Simulation Engineer to join our Physical AI practice, focused on building next-generation physics simulation capabilities for industrial clients. The ideal candidate brings strong hands-on expertise in computational fluid dynamics and thermal simulation, combined with applied proficiency in Physics AI techniques - Reduced Order Modeling (ROM), surrogate modeling, and Physics-Informed Neural Networks (PINNs), to transform high-fidelity simulation outputs into fast, deployable AI models at client scale.

In this role, you will run parametric CFD and continuous process simulations, build Physics AI pipelines that convert simulation data into real-time predictive models, and contribute to a tool-agnostic platform that serves clients across industries.

Roles and Responsibilities:

Architect simulation solutions across multi-physics domains (mechanical, thermal, fluid, electrical), with orchestration of cross-domain and discrete-event simulations.

Lead the development and execution of advanced CFD, multiphysics, and electromagnetic/electro-thermal simulations at both component and system levels, including turbulence, multiphase flows, conjugate heat transfer, fluid-structure interaction, and electromagnetic field modeling. Ensure accurate physics representation, coupling strategies and scaling workflows through automation, HPC, and cloud deployment.

Implement cross-domain and hybrid simulations, combining discrete-event modeling for factory workflows with physics-based models for material, fluid, and thermal processes.

Design and execute parametric simulation studies using structured Design of Experiments (DOE), sampling strategies (Latin Hypercube, Sobol), and output extraction protocols to generate Physics AI training datasets.

Develop and integrate ROMs, Surrogate models and Physics-Informed Neural Networks (PINNs) for fluid and thermal problems embedding governing equations (Navier-Stokes, heat equation, species transport) as physics constraints in neural network training loss, enabling physically consistent predictions from sparse simulation data.

Develop virtual sensor and derived variable models: extract ROM field predictions at specific spatial locations to simulate real-time sensor readings. Compute derived engineering KPIs, including remaining useful life (RUL), energy efficiency, yield uniformity from ROM outputs using physics-based and data-driven methods.

Champion co-simulation architectures, surrogate modeling, and tool integration across commercial software (ANSYS, COMSOL, Abaqus) and scripting environments (Python, MATLAB, Modelica).

Provide technical leadership and mentorship, guiding engineering teams in scalable simulation, best practices and architectural patterns.

Actively lead and participate in sales pursuits and pre-sales engagements, with responsibility for new project wins. This will involve multiple discussions and presentation of our capability and solutions to potential clients and other Accenture stakeholders/leaders.

Active participation and leadership in driving industry related thought leadership activities like patents, authoring whitepapers/POVs, proof of concepts, conference/academia presentations, capability presentations etc.

Work with global clients and cross-functional teams in a multicultural and distributed environment.

Professional and Technical Skills:

Must Have Skills:

Hands-on CFD simulation expertise with at least one primary solver - Ansys Fluent, OpenFOAM, or STAR-CCM+ covering mesh generation, boundary condition setup, turbulence modeling (k-e, k-w SST, LES), multiphase flow (VOF, Eulerian), conjugate heat transfer, and species transport for industrial-scale problems.

Strong numerical foundations in finite volume method, discretisation schemes, pressure-velocity coupling (SIMPLE, PISO), solver stability, and mesh quality assessment with the ability to diagnose and resolve convergence issues independently.

Applied knowledge of Physics AI methods: Reduced Order Modeling (ROM) using POD/SVD, surrogate modeling (Gaussian Process, Gradient Boosting, Neural Network), and DOE sampling strategies (LHS, Sobol, CCD) with the ability to connect CFD simulation outputs to Physics AI training pipelines.

Proficiency in Python for simulation automation and Physics AI development: NumPy, SciPy, Pandas, scikit-learn, PyTorch, h5py (HDF5 storage), Matplotlib, and PyVista (3D field visualisation). Familiarity with PyFluent, PyFOAM, or equivalent CFD Python APIs for programmatic solver control and result extraction.

Ability to extract, structure, and manage large field datasets (full nodal arrays: velocity, pressure, temperature) from CFD solver result files - cas/dat, VTK, RST, HDF5 formats - for Physics AI model training.

Creating digital twins, performing physics-based simulation, rendering RTX environments, or synthetic data workflows for real-time industrial applications.

Cloud engineering background, including automation, orchestration, and deployment of simulation pipelines in scalable environments.

Experience with manufacturing, industrial automation, automotive, or aerospace sectors.

Good to Have Skills:

Exposure to Ansys Twin Builder, FMI/FMU standard, or digital twin integration - deploying ROM as an FMU for system-level simulation or real-time inferencing in an operational environment.

Experience in scripting and automation, e.g. Python, C++, MATLAB, Modelica.

Background in digital twin development for factory planning, robotics simulation, or predictive maintenance.

Knowledge of NVIDIA Omniverse and Isaac Sim.

Familiarity with FMI/FMU, OpenModelica, OpenUSD, or robotics simulation standards.

Exposure to AI/ML-driven simulation workflows, Physics-informed ML, surrogate modeling, generative AI tools for model generation and behavior prediction.

Additional Info:

This position is based at Bengaluru, Pune

Candidate should have 8-10 years of experience

Bachelor s or master s in engineering, Physics