Data infrastructure that stays up.
I move millions of records a day across healthcare, insurance, and retail. Pipelines that run fast, stay clean, and do not page anyone at 3am.
Outcomes, not just
job descriptions.
From raw events to a
number someone trusts.
Every metric on this page rides through a shape like this one. Tap any stage to see what runs there and why it matters.
I make data move,
then make it trustworthy.
I have spent five years building the plumbing behind dashboards people actually depend on. Claims systems at a health plan, fraud and risk data at an insurer, and real time retail analytics at scale. The fun part is not the tool list. It is taking something messy and making it land on time, every time.
My default is boring in the best way. Clear contracts, tests that catch problems before a stakeholder does, and pipelines that recover on their own instead of waking someone. I care about the person three desks over who needs the number to be right.
A pipeline exists so a human can trust a number. That is the spec.
Tests and contracts catch the bad row before it reaches a report.
If a job can heal itself, it should. Nobody should babysit a DAG.
Cheaper compute and tighter storage are real wins, not afterthoughts.
A stack chosen for
uptime and clarity.
Where each tool sits
in my toolkit.
Adopt is what I reach for by default. Trial is in real projects. Assess is on my bench. Hover a blip to read why.
Five years, three
industries, zero drama.
Pipelines I am proud of.
The view I keep open
on a second screen.
A mock of the kind of control panel I wire up for every platform. Throughput, latency, and job health at a glance, refreshing in real time.
platform.health / production
Prefer a terminal?
Be my guest.
Try help, whoami, skills, experience, projects, certs, contact, or clear.
Credentials, kept current.
Have a pipeline
that needs to last?
I am looking for senior data engineering roles where reliability and scale actually matter. If that sounds like your team, let us talk.