Available for contract & consulting
I build cloud platforms and carry the pager for them.
Fourteen years of AWS, Kubernetes, Terraform, and CI/CD, most of it on call. The platform I run today peaks above 200 million requests a day. When a team needs a senior owner on an infrastructure problem for a while, that's the work I take.
200M+
requests/day at peak on the platform I run now
~$4M/yr
cloud spend held flat while traffic grew
$350K/yr
saved moving one transcode fleet to Spot
14+ yrs
building and running production infrastructure
What I take on
all services →platform
Cloud platform architecture
Full AWS environment builds and rebuilds: VPC, EKS, ECS, MSK, RDS Aurora. Done under live production load, documented on the way out.
kubernetes
Kubernetes and EKS operations
ArgoCD, Helm, Karpenter, KEDA. Cluster upgrades, ECS-to-EKS migrations, and node fleets sized for the workload instead of the worst case.
delivery
Terraform and CI/CD at scale
Hundreds of resources migrated off CloudFormation. Pipelines on GitHub Actions with OIDC role assumption, a plan on every PR, and no stored keys.
cost
Cloud cost engineering
A $500K AWS bill cut to $200K. $350K a year moved to Spot. Attribution first, then the savings, then the controls that keep them.
ai ops
AI ops and readiness
Claude Code against a live platform of 50+ microservices, behind a create-only gate and a safety review that run before any terraform apply. The speed, with a human still approving prod.
Articles
all articles →-
The attribution problem, or why "just tag everything" fails
Tags are the standard answer to AWS cost allocation and they've never once been enough on their own. What a workable attribution layer actually looks like.
-
The state of CI/CD in 2026
GitHub Actions won by default, Jenkins won't die, and most pipelines are still slower and leakier than they need to be. An assessment from someone who runs this stuff.
-
The state of infrastructure as code in 2026
Terraform after the license change, what OpenTofu did and didn't become, why CloudFormation still runs more production than anyone admits, and what I'd pick today.
Have a platform problem that needs an owner?
Contract and consulting engagements, remote from Colorado. Anything from a short assessment to a multi-quarter build.