Machine Learning — PhD Intern (Modeling and Data Analysis)

Keysight Technologies, Inc. • Calabasas, CA

2d

About The Position

Keysight’s AI team develops adaptive engineering intelligence systems combining measurement, simulation, and machine learning. This internship focuses on neural surrogate models that predict engineering responses from simulation or measurement data at a fraction of the cost of repeated high-fidelity simulation or physical characterization. The intern will develop and evaluate models, investigate efficient adaptation as design problems change, and integrate models into an engineering ML runtime environment.

The work spans data generation and assembly, model architecture and output-representation choices, accuracy evaluation, and analysis of performance against realistic simulation or measurement budgets. In a second phase, the intern will study specialization of trained models using LoRA and related parameter-efficient adaptation methods. The intern will collaborate with AI researchers, simulation and measurement experts, RF designers, and runtime engineers.

The internship offers practical experience applying machine learning to physical engineering data, parameter-efficient adaptation under data scarcity, RF and high-speed concepts, and taking a research model from Python experimentation through C++ and LibTorch integration. Mentorship and access to engineering tools and realistic datasets are provided. Candidates must be enrolled in an accredited college or university as of September 2027.

Requirements

• Currently pursuing or recently completed a PhD in Machine Learning, Computer Science, Applied Mathematics, Electrical or Electronic Engineering, or a related technical field.
• Hands-on experience with multiple neural architectures, such as transformers, neural operators, graph networks, and CNNs, and sound judgment about which architecture suits a problem.
• Strong Python and PyTorch skills, including experience training and debugging neural networks beyond tutorial scale.
• Strong C++ skills and hands-on LibTorch experience implementing and integrating models into a production runtime.
• Solid understanding of transfer learning, fine-tuning, and parameter-efficient methods such as LoRA.
• Strong applied statistics and exploratory data analysis skills, including sensitivity analysis and dimensionality reduction, and ability to diagnose models from their errors rather than aggregate scores.
• Experimental discipline, including rigorous baselines, careful ablations, experiment tracking, and reproducible results.
• Willingness to learn the underlying physics rather than treating it as arbitrary numerical data.
• Familiarity with version control, modular code structure, and reproducible experiment pipelines.
• Experience with CMake and Git workflows for sizeable C++ projects, and comfort profiling and debugging C++ code.
• Ability to design evaluations that answer a question rather than merely produce a favorable number.
• Clear technical communication with machine learning researchers and engineering stakeholders.
• Must be enrolled in an accredited college or university as of September 2027.

Nice To Haves

• Background in RF, microwave, high-speed digital, or test and measurement engineering; familiarity with S-parameters, filters, or interconnect.
• Experience with EM simulation tools and scripting interfaces, such as Keysight ADS, RFPro, Momentum, or comparable tools, or with engineering measurement and test automation environments.
• Experience with surrogate modeling, operator learning, physics-informed machine learning, or scientific machine learning.
• Experience with uncertainty quantification, conformal prediction, or Gaussian process regression.
• CUDA, cuDNN, or TensorRT experience, or prior experience shipping a research model in C++.

Responsibilities

• Generate parametric engineering datasets using simulation and/or measurement sources, considering sampling strategy and numerical noise.
• Design, train, and evaluate neural surrogate models for engineering response prediction.
• Investigate how output representation choices affect accuracy, particularly across operating regions or frequency ranges where responses vary sharply.
• Establish rigorous baselines and quantify accuracy as a function of simulation or measurement budget.
• Evaluate LoRA and related parameter-efficient methods for adapting trained models to new problem settings, comparing them with full and partial fine-tuning.
• Develop calibrated uncertainty estimates to determine when predictions can be trusted or should be deferred.
• Verify physical consistency and engineering constraints, including passivity, causality, and reciprocity where applicable, and report engineering-relevant metrics including worst-case behavior.
• Analyze simulation data and model behavior, including parameter sensitivity, response dimensionality, data quality, solver noise, and where prediction error concentrates across frequency and design space.
• Prototype model training and inference integration in C++ and LibTorch within an engineering ML runtime environment.

Benefits

• Mentorship across machine learning, simulation, measurement, and RF design.
• Access to engineering tools and realistic engineering datasets.
• Opportunity to apply modern machine learning to physical engineering data in an industrial setting and develop practical expertise in parameter-efficient adaptation, transfer learning, RF and high-speed concepts, and simulation workflows.

What This Job Offers

• Job Type: Full-time
• Career Level: Intern
• Education Level: Ph.D. or professional degree