One engineer.
Five production
systems.
I run the entire data and technology function for a multi-brand retail operator. Warehouse, orchestration, cloud services, internal ops applications, and the e-commerce storefront. Not five teams with five roadmaps. One person shipping them as one product.
Before this: Fortune 500 consulting at IBM, and a marketing data warehouse at WP Engine.
What I own, end to end
Every box below is a production system with real users and real money moving through it.
BigQuery warehouse, 400+ dbt models, Dagster orchestration, 20+ source systems.
Event-driven ingestion and rewards engines on Cloud Run, Cloud Functions, and Pub/Sub.
The company's BI layer, serving Retail, Franchise, and Finance on streaming APIs.
In-house loyalty, referrals, and an evergreen rewards program on shared coupon rails.
Shopify Hydrogen e-commerce, shipping through review by an external agency team.
Selected work
Each of these was designed, built, shipped, and is operated by one person.
225 declared sources
The data backbone for a multi-channel retailer, run by one person.
400+ dbt models over 225 declared sources, Dagster orchestration on 22 schedules, and event-driven ingestion from 20+ source systems. Intraday revenue used to wait for the next day's ERP sync. It now lands on a 10-minute loop.
+ vendor contract retired
Took loyalty away from a SaaS vendor without a customer losing a point.
150k+ members earning in real time at the register. Coupons minted automatically on a sub-30-minute loop from BigQuery balances. The warehouse, not an application, is the authority on what a customer is owed, so every balance is auditable back to source.
with no manual steps
An evergreen rewards program that runs itself off hardware in the store.
1.2M+ device scans ingested and 54k+ enrollments. A scan opens a 28-day window, a second scan closes it, and the reward is computed and texted to the customer in seconds. Nobody approves anything.
seats retired
Replaced Power BI as the company's BI layer and retired hundreds of seats.
Live Sales on a 10-minute loop, Sales Journal with a store-activity heatmap, Inventory Movement, and Franchise KPIs with projected Daily Need. Backed by streaming API endpoints on a dbt, Dagster, and Postgres stack.
Experience
Enterprise consulting, then data engineering at scale, then sole ownership.
Defyned Brands
2021 – PresentPrincipal Engineer
Sole engineer across the data platform, internal operations applications, GCP cloud services, and the Shopify Hydrogen storefront. Built an in-house loyalty platform that retired a vendor contract and cut six figures of points liability, and an event-driven rewards engine covering 1.2M+ device scans and 40k+ lbs of verified body-composition change. On the internal side: a streaming-API operations app serving Retail, Franchise, and Finance, and NewStore POS to Power BI reporting with row-level security across roughly 30 hierarchical roles. Storefront work ships through review by an external agency development team.
WP Engine
2018 – 2021Sr. Data Engineer
Marketing data warehouse, web analytics, A/B testing strategy. Built full-funnel attribution and led statistically sound experiment design.
IBM
2010 – 2018Consultant Team Lead, Sr. Solution Consultant
Led a team of services consultants for IBM's SaaS e-commerce and analytics platforms. Primary technical point of contact for Fortune 500 clients through complex implementations, and technical advisor to sales, support, and product management.
Coremetrics
2008 – 2010Implementation Engineer
Implementation training, technical consulting, and analytics best practices for Coremetrics Web Analytics. Level 2 support for client tagging and troubleshooting. Acquired by IBM in 2010.
Stack
What the systems above are actually built on.
Side projects
Built outside work hours, shipped and running.
PickCrunch
Create groups with your friends and pick NFL, NCAAFB, NBA, NCAABB, MLB and NHL games. Pick spreads and totals in matchup, ranked, or flat pick modes.