Scientific computing, data analysis, and process control
NumPy, SciPy, and Pandas for scientific computing and data analysis. Matplotlib for visualization. Used in semiconductor data work, the DUV energy deposition project, and the SPECTRA-Lab platform.
scikit-learn for classification, regression, clustering, feature engineering, and model validation on engineering and manufacturing data.
Python for automation, scripting, and instrument SDKs in semiconductor characterization workflows. Object oriented design for reusable analysis libraries.
MATLAB for device-level modeling alongside Sentaurus and Silvaco TCAD work. Used in the AlGaN/GaN HEMT electrothermal modeling project.
Numerical work for the DUV energy deposition project (Monte Carlo and FFT convolution model comparison) and signal processing in the undergraduate EIT and quantum memory research at UConn.
Completed coverage of supervised and unsupervised learning, feature engineering, model validation, and predictive modeling. Foundation for applied ML on engineering and manufacturing data.
Working knowledge of neural network and deep learning fundamentals: architecture choices, training and validation, and where they fit in the broader ML toolkit.
X-bar/R, I-MR, EWMA, CUSUM control charts, Western Electric rules, and Cp/Cpk/Pp/Ppk capability analysis. Implemented in the SPECTRA-Lab platform's SPC engine.
Design of Experiments (DOE), ANOVA, and regression analysis for process characterization and optimization.
8D root cause workflows, FMEA, and Pareto analysis. Applied at ASML on production-floor root cause investigations and at General Dynamics Electric Boat on weld and structural inspection findings.
Six Sigma Green Belt certified. Process improvement methodology and quality management system principles.