Machine Learning PhD Intern (Modeling and Data Analysis)

About the role#

This internship focuses on building machine learning surrogate models for engineering simulation and measurement. High-fidelity simulations and physical measurements often dominate the design and test loop for RF and high-speed engineering. You will develop models that predict engineering responses at a fraction of the cost, investigate how these models adapt as design problems shift, and integrate your work into an engineering ML runtime environment.

What you'll do#

• Generate parametric engineering datasets using simulation or measurement sources, accounting for sampling strategy and numerical noise.
• Design, train, and evaluate neural surrogate models to predict engineering responses.
• Investigate how output representation affects accuracy across operating regions or frequency ranges.
• Establish rigorous baselines and quantify accuracy based on simulation or measurement budgets.
• Evaluate LoRA and other parameter-efficient methods for specializing trained models to new problem settings.
• Develop calibrated uncertainty estimates to ensure predictions can be trusted or deferred appropriately.
• Verify physical consistency and engineering constraints, including passivity, causality, and reciprocity.
• Analyze simulation corpora and model behavior, including sensitivity, dimensionality, and error concentration.
• Prototype model training and inference integration in C++ and LibTorch within an engineering ML runtime environment.

What you'll need#

• Currently pursuing or recently completed a PhD in Machine Learning, Computer Science, Applied Mathematics, Electrical Engineering, or a related field.
• Hands-on experience with neural architectures including transformers, neural operators, graph networks, and CNNs.
• Strong skills in Python, PyTorch, C++, and LibTorch.
• Solid grasp of transfer learning, fine-tuning, and parameter-efficient methods like LoRA.
• Strong applied statistics, exploratory data analysis, sensitivity analysis, and dimensionality reduction skills.
• Experimental discipline, including rigorous baselines, ablations, and reproducible results.
• Ability to diagnose models from errors rather than aggregate scores.
• Experience with version control, modular code, and CMake/Git workflows for large C++ projects.
• Clear technical communication skills for researchers and engineering stakeholders.
• Willingness to learn the underlying physics of the data.
• Preferred: Background in RF, microwave, high-speed digital, or test and measurement engineering; experience with Keysight ADS, RFPro, Momentum, or similar EM simulation tools; familiarity with uncertainty quantification, CUDA, or scientific machine learning.

Location & details#

• Location: Calabasas, California.
• Term: Rolling.
• Modality: On-site.
• Enrollment: Must be enrolled in an accredited college or university as of September 2027.
• Compensation: This is a paid internship.