Dynamic Laboratory Assistant at Dartmouth-Hitchcock Medical Center, adept in data analysis and proficient in Python and MySQL. Successfully streamlined document management and inventory tracking, enhancing operational efficiency. Strong communicator with a proven ability to produce audit-ready reports, leveraging machine learning techniques to drive insights and improve processes.
Spam Email Classifier Project:, 1. Built a TF–IDF vectorization pipeline and trained a Logistic Regression model in Python, achieving 100% precision and recall on a 200-message test set., 2. Managed the full ML lifecycle—including data preprocessing, feature engineering, model training, and evaluation—using scikit-learn and pandas., 3. Packaged inference logic into a reusable `predict_spam()` function, documented workflows in Jupyter notebooks, and version-controlled all code with Git., Heart Disease Prediction Project:, 1. Engineered a full ML pipeline in Python—loading/cleaning the UCI heart.csv dataset, label-encoding categoricals, and selecting key features via correlation analysis—to predict heart disease with up to 88% accuracy., 2. Trained and evaluated both Logistic Regression and Random Forest models, standard-scaling inputs and reporting precision, recall, and F1-score for each class, then persisted models and preprocessors as `.pkl` files., 3. Built an interactive Gradio web interface for live patient-metric inputs and on-the-fly disease risk prediction, enabling non-technical stakeholders to explore model results instantly.