The infrastructure I operate. Kubernetes and OpenShift in production: resource configs tuned against live traffic, sustaining availability SLOs across the service mesh at Capital One. Docker as the portable unit of deployment.

Datadog is the observability stack I stood up team-wide — RUM, APM, logs, custom dashboards, availability alerts, and SLOs across every owned service. Instrumentation is not a phase of a project; it's the thing that tells you whether the project is working.

AWS across Lambda, S3, EC2, EMR, CloudWatch, and API Gateway, from academic research through production workloads. GitHub Actions and Vercel for everything outside the bank. CI/CD as the discipline of making deployment boring.