AI Engineer Scientific Machine Learning

About the Role

We are looking for an AI Engineer with strong expertise in Scientific Machine Learning (SciML), Neural Operators, PDE Learning, and Physics-Informed AI. The ideal candidate will have hands-on experience building machine learning models for scientific and engineering problems involving simulations, surrogate modeling, and computational physics.

Key Responsibilities

• Develop and deploy Scientific Machine Learning (SciML) solutions for complex engineering and simulation problems.
• Design and implement Neural Operator architectures such as Fourier Neural Operators (FNO) and Physics-Informed Neural Operators (PINO).
• Build surrogate models for accelerating physics-based simulations and numerical solvers.
• Work on PDE (Partial Differential Equation) learning and operator-learning frameworks.
• Apply Physics-Informed AI approaches to real-world scientific and industrial use cases.
• Collaborate with research and engineering teams to optimize model performance and scalability.
• Analyze model convergence, training behavior, and generalization across varying conditions.
• Leverage GPU-accelerated computing environments for large-scale model development.

Required Skills

• Strong programming skills in Python.
• Hands-on experience with PyTorch, TensorFlow, or JAX.
• Experience with Scientific Machine Learning (SciML).
• Knowledge of Neural Operators and operator-learning concepts.
• Hands-on expertise in at least one of:
• Fourier Neural Operators (FNO)
• Physics-Informed Neural Operators (PINO)
• Physics-Informed Neural Networks (PINNs)
• MeshGraphNet
• Understanding of:
• Partial Differential Equations (PDEs)
• Numerical Methods
• Surrogate Modeling
• Scientific Computing
• Computational Physics
• Experience working with GPU/HPC environments.

Preferred Skills

• Experience with DeepXDE.
• Experience with NVIDIA Modulus.
• Exposure to Computational Fluid Dynamics (CFD), Digital Twins, or simulation platforms.
• Research experience in Physics-Informed AI, Neural Operators, or PDE-based learning systems.

What We're Looking For

Candidates should be able to demonstrate:

• Clear understanding of Neural Networks vs Neural Operators.
• Knowledge of inputs and outputs of Neural Operators.
• Practical understanding of when to use FNO vs PINO.
• Strong grasp of PINO architecture and training methodology.
• Ability to troubleshoot model training challenges, including convergence and loss optimization.
• Experience applying AI to scientific, simulation, or physics-based domains.