Delivery
Big moves broken into waves small enough to roll back in an hour.
- Migration waves planned by risk
- Rehearsed cutovers and rollbacks
- One weekly status anyone can read
Head of cloud delivery
I lead cloud delivery at Halcrest Mutual, where 120 engineers move core insurance systems off aging data centers without stopping the business. Fourteen years of migrations, platform teams and calm on-call rotations.
From systems engineer to running a cloud organization across three industries.
Data center exits and platform moves in insurance, logistics and healthcare.
Platform, SRE and migration teams across Denver, Austin and two remote pods.
Held on customer-facing services through two full peak seasons in a row.
Weekly releases became daily once pipelines and rollbacks were automated.
Fourteen years, four companies and one thread: moving critical systems without anyone noticing the move.
2021 to now
Halcrest Mutual
Run the cloud organization for a regional insurer: migration, platform engineering and site reliability, 120 engineers across five teams.
2017 to 2021
Copperline Freight
Led the move of tracking, dispatch and warehouse systems to the cloud and built the first SRE team.
2014 to 2017
Bluestem Health Network
Ran infrastructure programs for a network of eleven hospitals, including its first data center exit.
2012 to 2014
Ironvale Data Services
Built and patched Linux and virtualization estates for managed hosting clients while carrying the pager.
Five programs, each with a clear problem, a plan and a measured result.
Insurance
Moved 140 claims and policy applications out of two aging data centers without a single claims outage.
Both data centers closed four months early and claim intake got 38% faster.
Insurance
Teams waited three weeks for a new service environment and every team built pipelines its own way.
New services go live in two days instead of three weeks. 80% of teams now use the paved road.
Logistics
Tracking went down for hours each holiday peak, right when shippers and customers needed it most.
99.95% uptime through two peak seasons and 60% fewer high-severity incidents.
Logistics
Eleven regional warehouses ran eleven versions of the same inventory system, each patched by hand.
One shared system for all eleven sites and 45% fewer support tickets in the first year.
Healthcare
A flood-prone basement data center held lab, imaging and scheduling systems for eleven hospitals.
Zero unplanned clinical downtime and a clean compliance audit after the move.
How I run a cloud organization: small accountable teams, plain metrics and no heroics.
Big moves broken into waves small enough to roll back in an hour.
Reliability treated as a product with owners, targets and a backlog.
Leads who grow leads, so the organization does not depend on one person.
Guardrails built into the platform instead of review boards and long forms.
Planning a data center exit or a platform team? I am glad to compare notes over a short call.
Email jonathan@example.com