AI Enabled Device Modeling

GlobalFoundries•Essex, VT

1d•$58,400 - $100,800

About The Position

GlobalFoundries is seeking a new graduate with strong artificial intelligence and machine-learning skills to develop AI-enabled methods for semiconductor device modeling. The engineer will combine device physics, measured silicon data, simulation data, and modern AI/ML techniques to accelerate model extraction, improve predictive accuracy, and create scalable modeling workflows for CMOS, cryogenic, photonic, superconducting, and other emerging devices. The role involves developing and evaluating machine-learning methods for device modeling, building reproducible data pipelines, validating models, automating Python-based workflows, and collaborating across device, characterization, compact-modeling, PDK, EDA, and circuit-design teams. The engineer will document methods and results and help transition successful proofs of concept into maintainable device-modeling and design-enablement workflows.

Requirements

Bachelor's or Master's degree completed or expected in Electrical Engineering, Computer Engineering, Computer Science, Data Science, Applied Physics, Physics, or a related field from an accredited degree program.
Academic or project experience in machine learning, deep learning, statistical modeling, optimization, or scientific data analysis.
Strong Python programming skills and experience with common scientific and machine-learning libraries such as NumPy, pandas, SciPy, scikit-learn, PyTorch, TensorFlow, or JAX.
Understanding of supervised learning, regression, neural networks, model validation, overfitting, uncertainty, and quantitative performance evaluation.
Foundational understanding of semiconductor device physics and transistor electrical behavior, with interest in applying AI methods to device modeling.
Experience working with structured scientific or engineering datasets, including data preparation, visualization, feature engineering, and reproducible analysis.
Ability to communicate technical results clearly and collaborate across AI, device, modeling, characterization, and design teams.
Fluency in written and spoken English.

Nice To Haves

Graduate coursework, thesis research, internship, co-op, or substantial project work in AI/ML for semiconductor devices, electronic design automation, scientific machine learning, or computational physics.
Experience with physics-informed neural networks, differentiable programming, Bayesian optimization, Gaussian processes, active learning, or uncertainty quantification.
Exposure to compact transistor models, SPICE simulation, parameter extraction, TCAD, cryogenic device behavior, photonic devices, superconducting devices, or quantum hardware.
Experience using experiment tracking, version control, testing, or reproducibility tools such as Git, MLflow, Weights & Biases, containers, or automated workflows.
Experience with Linux environments, high-performance computing, GPU acceleration, cloud computing, or large-scale simulation and training datasets.
Ability to balance predictive performance with physical consistency, interpretability, robustness, and computational efficiency.
Demonstrated technical initiative through publications, conference presentations, open-source contributions, competitions, or interdisciplinary research projects.
Project management skills, including the ability to innovate, execute solutions that matter, and navigate ambiguity.
Strong written and verbal communication skills and strong planning and organizational skills.

Responsibilities

Develop and evaluate machine-learning methods for semiconductor device modeling, including parameter extraction, surrogate modeling, model calibration, and prediction of device characteristics.
Build reproducible data pipelines combining measured device data, compact-model simulations, TCAD results, and relevant process or test metadata.
Train, validate, and compare AI/ML models using appropriate baselines, holdout datasets, error metrics, uncertainty analysis, and physics-based constraints.
Apply neural networks, Gaussian processes, optimization, sensitivity analysis, and physics-informed machine learning to device-modeling problems.
Automate model fitting, feature extraction, hyperparameter optimization, regression testing, visualization, and reporting using Python-based workflows.
Work with device, characterization, compact-modeling, PDK, EDA, and circuit-design teams to translate technical requirements into usable AI-enabled modeling solutions.
Document datasets, assumptions, model architecture, training methods, validation results, limitations, and recommended usage to support traceability and knowledge transfer.
Contribute to proof-of-concept demonstrations and transition successful methods into maintainable device-modeling and design-enablement workflows.
Perform all activities safely and responsibly and support Environmental, Health, Safety & Security requirements and programs.
Work and collaborate on other projects or assignments as needed.