Mahmoud Elhadidy

Scientific Machine Learning Research Assistant @Scientific Computing And Imaging Institute At The University Of Utah

Salt Lake City, UT, US

Led the implementation of advanced ML algorithms to tackle inverse and forward problems in mechanical simulations and physics, achieving 94% accuracy in constructing physical quantities of interest, and driving significant improvements in simulation accuracy and computational efficiency. Pioneered the exploration and integration of Large Language Models (LLMs) and foundation models such as transformers to enhance simulations, achieving 70% savings in setup and running time while significantly improving computational efficiency and reducing costs. Spearheaded the application of compressed sensing (CS) and model reduction techniques to streamline mechanical simulations, reducing data complexity by 92%. This initiative boosted computational efficiency while maintaining accuracy, driving significant process improvements Led the implementation of CNNs with physical constraints to enhance low-resolution simulations, reducing computational costs by 80% and significantly cutting simulation time compared to high-fidelity methods, driving efficiency improvements across projects.**Tools and Skills:**• Programming Languages: Python• ML/DL Frameworks: PyTorch, TensorFlow, Keras• Data Science Libraries: pandas, NumPy, scikit-learn, SciPy• HPC and Parallel Computing: MPI, OpenMP• CFD and FEA Tools: FEniCS, SimVascular, ANSYS• Data Visualization Tools: ParaView, Python (Matplotlib, Seaborn)• Version Control: Git, GitHub• Problem-Solving in Programming, Data Structures, and Algorithms

University of Utah

Doctor of Philosophy - PhD, Scientific Machine Learning, Mechanical Engineering

Mansoura University

Bachelor of Mechatronics Engineering, Mechactronics Engineering

University of Utah

Master of Science - MS, Mechanical Engineering, Data Science