AI/ML Specialist – Surrogate Model Creation (Jet & Cryogenic Engines)

Scope of Work

We are seeking a highly motivated AI/ML Specialist to develop fast, accurate surrogate models for multidisciplinary simulation and optimization workflows in high-performance turbomachinery. This includes modeling complex physics of jet and cryogenic engines using reduced-order modeling, Gaussian processes, neural networks, and symbolic regression techniques. You will work closely with CFD, thermal, structural, and control teams to accelerate design iterations and enable real-time digital twin deployment.

Nature of Work

• Build physics-informed machine learning models based on CFD/FEM datasets
• Create surrogate models for turbomachinery components: blades, nozzles, combustors, and turbopumps
• Deploy model training pipelines using TensorFlow, PyTorch, and Scikit-learn
• Perform data fusion and feature engineering for sparse, multi-fidelity datasets
• Integrate AI models into digital twin platforms and design optimization loops
• Collaborate with aerospace domain experts to ensure physical validity
• Support explainable AI development and verification for mission-critical applications

Job Requirements

• B.Tech/M.Tech/M.S./PhD in Aerospace, Mechanical, Computer Science, Data Science, or Applied Mathematics
• Strong programming background in Python, with experience in ML libraries (PyTorch, TensorFlow, Scikit-learn)
• Hands-on experience in developing surrogate models for engineering simulation or design optimization
• Understanding of CFD, FEM, or multiphysics simulation data structure
• Experience with symbolic regression, Bayesian optimization, or meta-modeling techniques
• Working knowledge of HPC environments and GPU acceleration (preferred)
• Clear documentation skills and ability to present AI models to non-ML stakeholders

Eligibility

• Indian nationals or OCI cardholders eligible to work in India
• Passion for building AI tools for aerospace and defense applications
• Research scholars and early-career professionals with publications or open-source contributions in relevant areas are encouraged to apply