I build and govern production analytics systems in regulated healthcare — pipelines, semantic layers, and data products that clinical, operational, and executive teams can trust and act on.

I started as a news writer and book editor, then moved into healthcare research and analytics. That background still shapes how I work: clear questions, defensible methods, and results that change decisions.

Currently: Lead Data Engineer at Baltimore Health Analytics, working on Medicare Advantage quality and performance analytics.

• Production data pipelines — EHR data modeling across Epic, Cerner, Veradigm, and athenahealth; ETL/ELT on AWS and Databricks; stored procedures built to survive edge cases in regulated populations
• Analytics governance — Semantic layers, metric definitions, and dbt-versioned models that give stakeholders one version of truth
• Value-based care analytics — ACO, MSSP, HEDIS, and Medicare Stars performance tracking; benchmarking and significance testing against CMS methodology
• Platform reliability — Observability, incident response, and cost optimization across cloud data infrastructure

Payer Quality Analytics Platform — Lead data engineering for a SaaS Medicare Advantage analytics platform. Designed stored procedure architecture for regulatory measure calculation, CMS specification validation, and plan-level performance reporting. (MariaDB, Python, Ruby on Rails)

Embedded Refills and Care Gaps — Designed the data model and daily cohort refresh pipeline; standardized legacy SQL into shared stored procedures (~70% reduction in codebase), improving maintainability and cross-customer deployment. (AWS Redshift, dbt, Airflow, FHIR, Tableau, Power BI)

Revenue and Program Performance — Built benchmarking and trend tracking for ACO/MSSP/HEDIS/Stars programs; reduced storage costs ~50% and ETL load time by 24+ hours through infrastructure redesign; governed dashboards enabled 7× user growth and eliminated 400+ manual hours per quarter. (S3, Redshift, dbt, SQL, Python, Tableau)

Published on accountable care organization shared savings, clinic design and team efficiency, and EHR optimization in primary care. Presented at state and national conferences on population health, community-engaged mental health research, and quality and safety improvement — audiences ranging from clinical teams to health system leadership.

Data and cloud: SQL, Python (pandas, PySpark, Jinja, boto), R, dbt, Airflow, Git, AWS (S3, Redshift), Databricks, Perl, Ruby
BI and visualization: Tableau, Power BI, QuickSight, Sisense for Cloud Data Teams
Observability: Grafana, DataDog, SumoLogic
Healthcare data: Epic Clarity, Cerner, Veradigm/Allscripts, athenahealth, FHIR, HL7, CMS technical specifications, Public Use Files