Data Science & Analytics student and ML Practitioner. Turning raw data into answers people can actually use.
These days I build ML models and statistical analyses in Python, R, and SQL.
I'm finishing my M.S. in Data Science & Analytics at Grand Valley State University. My path here ran through three years at DXC Technology, where I automated reporting and built Power BI dashboards for an enterprise SAN environment,
and an internship at Isuzu Motors, where I learned that a good Pareto chart can settle an argument faster than a meeting.
What keeps me hooked is the same in every job, finding patterns in data that change what someone decides to do.
A 24/7 chat assistant for GVSU's Language Resource Center. Non-technical staff update the knowledge base in Google Sheets; the pipeline grounds every answer in it — no hallucinated policies, no robotic tone.
End-to-end similarity search over 10,000+ images: ResNet18 embeddings, a FAISS IVF-Flat index, and an interactive Streamlit UI — fully containerized with Docker.
A 93-day recursive forecast using a delta-modeling strategy to prevent error compounding, with 72 engineered features and an inverse-RMSE weighted ensemble of five models.
Logistic and Poisson GLMs on U.S. Census education data quantifying which socioeconomic factors predict student performance — tuned with AIC selection and 5-fold cross-validation.
Integrated 67,000+ incident records with Census demographics to map geographic and socioeconomic patterns across all 50 states, validated with Monte Carlo permutation testing.
Chronic-issue analysis at Isuzu Motors: Pareto ranking of defects by frequency and cost, fishbone root-cause tracing, and corrective actions validated on the production line.
Python, R, SQL, MATLAB
Pandas, NumPy, Scikit-learn, LightGBM, PyTorch, FAISS, tidymodels
Power BI, Tableau, Streamlit, Plotly, ggplot2, Observable
MySQL, MongoDB, Redis, Databricks, AWS (Certified Cloud Practitioner)
Git, Jupyter, Docker, ServiceNow, Quarto, R Markdown, n8n
Statistical modeling, GLMs, time-series forecasting, ensembles, EDA, A/B testing, reproducible research
Statistical Modeling · Machine Learning · Data Engineering · Regression Analysis · Data Mining · Generative AI
Systems thinking, numerical methods, and structured problem decomposition — a foundation that transfers directly to data pipeline design and root-cause analysis.
Project deep-dives, statistical modeling walkthroughs, and lessons from building ML pipelines — written for readers, not just reviewers.