Data Analyst • Data Storytelling • Human-in-the-Loop ML • Automation • Business & Operations Management
With 15 years in business and operations management, I bring real-world leadership, financial reporting, and marketing expertise to my data work. I use SQL and Python to build narrative dashboards, run rigorous model-evaluation workflows, and document with stakeholder empathy.
My portfolio blends production-minded tooling —with synthetic data engines and case-study storytelling. Together, these projects simulate real-world stakeholder scenarios, validate workflows, and deliver production-ready code and insights.
A curated, modular knowledge base of checklists, decision cards, guidebooks, and reusable scripts across Python, SQL, and analytics workflows. Originally built in Obsidian and published for quick, skimmable reference.
Modular ETL/diagnostics for Jupyter + CLI. Data validation, outlier detection, duplicate handling, and clean exports with robust logging.
YAML‑driven evaluation CLI and notebook suite featuring visualizations, SHAP explainers, and exportable reports. Designed for reproducible benchmarking and production‑grade model validation.
Synthetic e‑commerce data engine for realistic transactions, customers, and products. Configurable schemas and volumes for repeatable analysis and demos.
A story‑driven SQL skill‑building suite featuring datasets from the E‑commerce Data Generator. Includes an end‑to‑end case study with analysis, dashboards, and deliverables.
A domain‑specific retelling of the Palmer Penguins case study using tools from the Dirty Birds Data Generator, Analyst Toolkit, and Model Evaluation Suite.
Exploratory analysis and feature engineering on a public dataset, with reproducible notebooks, visualizations, and ML Modeling.
The fastest way to reach me is email: me@garrettschumacher.com. You can also connect on LinkedIn.