After more than 15 years working in editing, publishing, communications, and design, including five years in management, I made a deliberate career transition into computer science and artificial intelligence. I completed an M.S. in Computer Science specializing in Artificial Intelligence and Machine Learning, earning two WGU Excellence Awards for my graduate AI/ML work and the AWS Certified Machine Learning Engineer – Associate certification.
I'm now focused on applying that combination of technical training, problem solving, communication, and professional experience to machine learning and AI engineering.
My portfolio includes work in deep learning, computer vision, natural language processing, ensemble learning, probabilistic graphical models, feature engineering, and algorithmic optimization.
Built an interpretable Bayesian Network for personalized movie-preference prediction using MovieLens ratings and Tag Genome data. Developed user-specific preference profiles and compatibility features while using a history/holdout design to prevent information leakage. The final six-feature network achieved 70.82% accuracy and a 0.7029 macro F1-score across more than 1.5 million holdout observations.
Technologies: Python, pgmpy, scikit-learn, pandas, Bayesian Networks, probabilistic graphical models
Developed and optimized a VGG-inspired convolutional neural network for CIFAR-10 image classification. The final model achieved 88.76% test accuracy and incorporated model evaluation, hyperparameter optimization, and class-level performance analysis.
Technologies: Python, TensorFlow/Keras, CNNs, computer vision, deep learning
Developed a HistGradientBoosting classification model to analyze and predict speed-dating decisions using relationship-based feature engineering. The project explores how individual characteristics, partner perceptions, and interpersonal differences contribute to dating outcomes.
Technologies: Python, scikit-learn, HistGradientBoosting, feature engineering, classification, model interpretation
Developed and compared machine learning approaches for predicting animal-control incident response times. Evaluated linear regression, random forest, XGBoost, and ensemble approaches, with XGBoost selected as the final model.
Technologies: Python, XGBoost, scikit-learn, pandas, regression, ensemble learning
Built an end-to-end NLP pipeline for classifying Amazon product reviews using linguistic preprocessing, TF-IDF, and logistic regression. Cross-validated hyperparameter optimization improved test accuracy from 72% to 77% and precision from 69.81% to 84.21%.
Technologies: Python, spaCy, scikit-learn, TF-IDF, NLP, sentiment analysis, GridSearchCV
Compared shortest-path algorithms for emergency medical dispatch optimization, including Dijkstra's and Floyd-Warshall algorithms, with an emphasis on computational tradeoffs and efficient routing.
Technologies: Python, graph algorithms, Dijkstra's algorithm, Floyd-Warshall algorithm, algorithm analysis
Machine Learning: Classification, Regression, Ensemble Learning, Feature Engineering, Hyperparameter Optimization, Model Evaluation
Deep Learning: TensorFlow, Keras, Convolutional Neural Networks, Computer Vision
NLP: spaCy, TF-IDF, Text Classification, Sentiment Analysis
Probabilistic Modeling: Bayesian Networks, Probabilistic Graphical Models
Languages & Libraries: Python, scikit-learn, XGBoost, pandas, NumPy
Tools & Platforms: AWS, Jupyter, Git, GitHub, VS Code
- Two-time WGU Excellence Award recipient for exceptional performance in graduate AI/ML projects
- AWS Certified Machine Learning Engineer – Associate
Before transitioning into AI and machine learning, I spent more than 15 years working in editing, publishing, communications, and design, including five years in management. That experience developed a strong foundation in analytical problem solving, quality control, communication, and leading complex projects—skills I now bring to machine learning and AI development.