AI Engineering Intern

Summary: Hexagon Asset Lifecycle Intelligence is offering a hands-on internship for students interested in applying advanced AI methods to real engineering problems. The internship focuses on designing, implementing, and training Physics-Informed Neural Networks (PINNs) to model stress, displacement, and thermal expansion behavior in complex piping systems. Responsibilities: - Convert mechanical/pipe‑stress code equations into PDE or variational forms for use in PINN models - Develop and train PINNs in PyTorch to model stress, displacement, and thermal expansion responses - Create and preprocess simulation data; use TensorBoard for experiment tracking and visualization - Build NN training/learning environments and agents to explore piping layout optimization strategies - Validate AI models against existing solver results and assist in integrating NN prediction results into analysis workflows Required Qualifications: - Currently pursuing a degree in a related technical field (Mechanical Engineering with computer science background preferred) - Solid understanding of mechanics, differential equations, and numerical methods - Proficiency with Python and strong interest in machine learning - Strong analytical and problem‑solving skills - Ability to collaborate effectively with cross-functional teams Preferred Qualifications: - Experience with PyTorch, PINNs, or scientific machine learning - Familiarity with FEA concepts, PDE solvers, or reinforcement learning - Knowledge of piping systems and codes such as ASME B31 is a plus - Strong software engineering fundamentals Required Skills: Physics-Informed Neural Networks, PyTorch, Differential equations Important Skills: Python, Numerical methods, Machine learning Nice-to-Have Skills: Analytical skills, Software engineering fundamentals, FEA concepts, PDE solvers, Reinforcement learning, piping systems, Problem-solving skills, Collaboration