Senior Applied Scientist – AI/ML Calibration & Optimization

KLA•Milpitas, CA

1d•$136,300 - $231,700

About The Position

KLA is seeking a Senior Applied Scientist to modernize calibration, diagnostics, and optimization for electron-optical systems. The role uses machine learning, mathematical and statistical modeling, physics-informed AI, and large-scale engineering data to reduce calibration time, improve accuracy and repeatability, compensate for drift, strengthen tool-to-tool matching, and enable intelligent diagnostics and recovery. The scientist will develop algorithms, validate them against hardware data, and collaborate with engineering teams to integrate successful solutions into products. The work spans electron sources, electron-optical columns, steering systems, detectors, alignment, focus, high-voltage, and vacuum-related performance. The role supports the full path from concept through implementation, experimental validation, product qualification, and production deployment.

Requirements

M.S. or Ph.D. in Physics, Applied Physics, Applied Mathematics, Statistics, Data Science, Electrical Engineering, Computer Science, Optical Engineering, or a related quantitative discipline.
Three or more years of relevant industrial or applied research experience; relevant doctoral or postdoctoral work may qualify when it demonstrates substantial ownership of applied modeling, algorithm development, experimentation, or engineering-data analysis.
Demonstrated experience developing and validating machine-learning, statistical, or mathematical models for complex physical, scientific, or engineering problems.
Strong foundation in several of the following: numerical optimization; probability and statistics; linear algebra and numerical methods; regression, estimation, or system identification; time-series or multivariate analysis; sensitivity analysis, uncertainty quantification, or error propagation.
Proficiency in Python and practical experience with a modern machine-learning or scientific-computing framework such as PyTorch, TensorFlow, JAX, scikit-learn, SciPy, or a comparable environment.
Experience working with substantial engineering, experimental, simulation, sensor, telemetry, or operational datasets.
Evidence of personally developing or materially improving an algorithm, model, or optimization method, rather than solely applying an off-the-shelf model.
Experience designing experiments, evaluating model performance, and correlating analytical or model predictions with measured results.
Ability to independently frame ambiguous technical problems, test hypotheses, evaluate tradeoffs, and communicate conclusions to cross-functional engineering partners.

Nice To Haves

Experience with calibration, system identification, controls, diagnostics, drift compensation, predictive maintenance, or condition monitoring.
Familiarity with electron-optical systems, SEM, charged-particle optics, semiconductor inspection, or metrology equipment.
Experience with advanced scientific instruments, imaging systems, robotics, autonomous systems, aerospace systems, medical imaging, industrial equipment, or other sensor-rich physical platforms.
Experience with fleet data, machine telemetry, high-dimensional time-series data, or distributed engineering-data environments.
Familiarity with physics-informed neural networks, neural operators, surrogate models, differentiable simulation, inverse problems, or hybrid physics/data models.
Experience with design of experiments, statistical process control, robust optimization, Bayesian methods, or uncertainty-aware modeling.
Experience accelerating scientific workloads with GPUs, HPC, or parallel computing.
Experience translating Python or research prototypes into production software; C++ experience is helpful but not required.
Experience contributing algorithms to a released product or operational engineering workflow.

Responsibilities

Redesign calibration and recalibration workflows to improve speed, accuracy, repeatability, robustness, and system productivity.
Develop and validate calibration algorithms for electron sources, electron-optical columns, deflectors and beam-steering systems, detectors, alignment and focus functions, and performance compensation associated with high-voltage and vacuum behavior.
Apply machine learning, statistical modeling, and physics-informed AI to predict calibration outcomes, model system behavior, and improve calibration decisions.
Develop methods to detect drift, anomalies, and performance excursions using engineering, operational, and fleet-level data.
Optimize calibration sequences to reduce time, eliminate unnecessary steps, and balance accuracy, throughput, and system risk.
Develop adaptive and self-correcting approaches that respond to changing system conditions and support more autonomous calibration behavior.
Create intelligent diagnostic and recovery methods to accelerate root-cause identification and restore system performance after calibration or hardware excursions.
Build and combine physics-based and data-driven models, including correlation and response models, sensitivity studies, uncertainty estimates, and error-propagation analyses.
Analyze large, heterogeneous engineering datasets to identify performance patterns, tool-to-tool differences, failure modes, and opportunities for improvement.
Design and execute experiments to evaluate calibration repeatability, reproducibility, robustness, and generalization across operating conditions.
Correlate model predictions with experimental measurements and use discrepancies to refine models, algorithms, and calibration strategies.
Partner with electron optics, systems, software, algorithm, manufacturing, and service teams to translate system requirements into deployable calibration and monitoring solutions.
Support work from concept through algorithm implementation, experimental validation, product qualification, and production deployment.
Evaluate advances in scientific machine learning, optimization, anomaly detection, surrogate modeling, uncertainty quantification, and intelligent control.