About
One engineer, and the reason that's the point.
Graham Cloud Services is how I take on SRE and observability contracts: seven years of keeping production systems running, now with AI doing the digging and me making the calls.
Where I work
Based in Durham, NC, working remotely with engineering teams across the United States, on AWS and Azure.
Background
7+ years working in DevOps and as a Site Reliability Engineer, including compliance frameworks like HIPAA, SOC2, and HITRUST. I've led an observability migration off Datadog to Grafana and Prometheus that cut $240K+ a year while keeping full coverage, run multi-environment EKS clusters at scale with Karpenter and GitOps via ArgoCD, rebuilt logging infrastructure to resolve compliance failures, and automated database credential rotation on RDS. Infrastructure as code, in Terraform, is how every change happens, whether it's for a large system or a small one.
Credentials
AWS Certified Solutions Architect, Associate, scored 903 out of 1000, valid through November 2028. Bachelor of Science in Computer Science, Brigham Young University.
Why I work this way
Reliability work is usually hired out to a firm with layers between you and the person doing it, or kept in-house on a team that is stretched thin. I offer a third option: direct access to someone who has run this kind of system before, joining your team for the job at hand.
AI is what makes one engineer go further. It reads the logs, metrics and traces faster than any person can, ties a spike to the change behind it, and drafts the runbook. It never gets the last word: I review and approve every change, inside your process, and I own the result.
How I work
Scoped in writing before anything starts. Your accounts, your access rules, your change process. Sensitive data such as PII and PHI stays masked before it reaches any tool. Replies within one business day, always from me, and sooner when something is down.
What I won't do
Sell you work you don't need, cut a corner that shows up as a failure later, lock you into a platform you can't leave, or quote a number that changes on the invoice.
Have a reliability problem worth 20 minutes?
Book a call, or send the capabilities sheet to whoever needs to see it first.