Some of my Projects
Disclaimer
Since my college days, I've been building projects out of curiosity and to explore different database platforms, programming languages, and technologies. I wanted to understand how they worked, compare their strengths and weaknesses, and keep learning by building. The one thing I didn't do very well was organize everything. My code ended up scattered across different computers, folders, and repositories, and plenty of projects never made it into a portfolio.
Then AI came along, and I realized I no longer had an excuse. It became a fantastic tool for helping me organize years of code, improve documentation, and finally bring my projects together into one professional portfolio. That's exactly what you're looking at now. AI is an incredible productivity tool, but it's just that—a tool. One thing I've learned is that AI will confidently agree with almost anything you tell it, so you have to challenge it, verify its suggestions, and know when it's wrong. The ideas, architecture, debugging, problem-solving, and final decisions behind these projects are still mine.
I have plenty more ideas I'd like to build out—they just need time, which is in short supply between a full-time job and three kids at home. If you're visiting this page and have ideas for how any of these projects could be improved, feel free to send me an email. I'll get to it. Fair warning that “soon” might mean a few weekends out.
Everything here uses publicly available data that was available at the time each project was built. Source code and documentation live in dedicated GitHub repositories linked from each project—except for the MAPS website and the iOS and Android applications. Those repositories remain private because they contain proprietary business logic, API integrations, configuration details, and user-related information. Keeping them private protects both the application and its users while avoiding unnecessary security and legal risks.
An end-to-end Databricks Lakehouse project investigating associations between U.S. cancer rates and environmental, lifestyle, and water quality factors. I ingested 52.8 million rows across 33 Delta tables, built a full Medallion Architecture pipeline, trained ML models (R²=0.81), and published 3 interactive Tableau dashboards.
Apache Spark · Delta Lake · Databricks · scikit-learn · Tableau Public View ProjectAn end-to-end Databricks medallion pipeline analyzing whether Medicare actually pays more for better hospital care. Built on 10 million CMS records across 13 Delta tables, the analysis finds spending and outcomes are nearly independent at the hospital level (r = 0.033).
Healthcare · Data Engineering · Cost-Quality Analysis View ProjectA data-driven dashboard analyzing the projected 10-year impact of the OBBB Act on Medicaid, Medicare, healthcare funding, coverage losses, and state-level outcomes.
Healthcare Policy · 10-Year Forecast · Dashboard Open DashboardThe MAPS Assessment is built on a proprietary framework that identifies whether two career paths secretly reinforce each other — or drain each other. It is a career intelligence platform built for people who have always been drawn to more than one professional path.
Career Intelligence · Dual-Path Assessment · Product Visit MAPS
An end-to-end PostgreSQL database administration project built on 9.6 million CMS Medicare records. Covers schema design, performance diagnostics, index optimization (4,136× improvement), backup and recovery, monitoring, role-based access control, row-level security, and audit logging.
PostgreSQL · Database Administration · Performance Tuning · Security View Project
An end-to-end MySQL database administration project built on publicly available EPA, CDC, and USDA datasets. Covers star schema design, bulk loading of 22.7 million rows, performance optimization (95.5% query improvement), backup and recovery with full restore verification, and role-based access control.
MySQL · Database Administration · Performance Tuning · Security View Project
An end-to-end database migration project built on a CMS Medicare provider dataset. Covers source assessment, schema conversion — including a custom audit-trigger rewrite — a custom Python ETL pipeline moving 20.5 million rows, three-layer validation, and honest performance tuning against the original PostgreSQL system.
PostgreSQL · SQL Server · Migration · Python ETL View Project
A hands-on portfolio project simulating the real-world responsibilities of an enterprise SQL Server DBA, built on 113.5 million rows of real NOAA climate data. Covers performance tuning, backup and recovery, security (TDE, RLS, masking), auditing and compliance, SQL Server Agent automation, Always On Availability Groups, and Extended Events monitoring.
SQL Server · Database Administration · High Availability · Monitoring View ProjectA data-driven analysis exploring how educational technology adoption influences student performance, learning outcomes, engagement, and equity across institutions — combining longitudinal data, ed-tech usage metrics, and outcome modeling.
EdTech · Outcome Analytics · Research View Analysis