DevOps, Observability & AI Engineering Consulting
Behind Roatán Websites is a senior platform engineer who builds observability systems and AI-assisted engineering tooling for a living. If your company needs that kind of work done, not a website, this is the page for that.
Who This Is
I'm Ricky McKenzie, currently Technical Lead for Observability & Platform Engineering at Viasat, where I own the org-wide GCP logging roadmap across 1,000+ projects, lead the migration of 700+ Grafana dashboards from Splunk to BigQuery, and run a production OpenTelemetry log pipeline at 5,000+ host scale. Before that, DevOps engineer work across Splunk, Terraform, Ansible, and AWS, and systems administration at Dexcom. Full details on LinkedIn.
What I Can Help With
- Observability builds: logging architecture, Splunk/Grafana/Prometheus, OpenTelemetry pipeline design (Bindplane receivers/processors/exporters), BigQuery-backed log analytics and cost optimization
- Migrations & modernization: moving off legacy tooling (Splunk, CloudFormation, Jenkins) onto GitOps-managed infrastructure, dashboard and query conversion at scale
- Software/tooling onboarding: CI/CD pipeline design (GitHub Actions), infrastructure as code (Terraform, Ansible), self-service platforms your teams can actually use
- AI-assisted engineering: custom Claude Code skills and agents built for real engineering workflows, anomaly detection pipelines, AIOps alerting, the practical version of "AI adoption" rather than the slideware version
What This Has Looked Like in Practice
1,000+
GCP projects under centralized logging governance
700+
Grafana dashboards migrated (2,078 panels remediated in one pass)
5,000+
hosts on a production OpenTelemetry log pipeline
$210K/mo
modeled savings identified from a log-exclusion strategy
All figures from current work at Viasat, not hypothetical.
Frequently Asked Questions
Is this the same as the website design services?
No, different thing entirely. The rest of this site is a small web shop building sites for island businesses. This page is about hiring me directly, as an individual engineer, for observability and platform work, usually remote, for companies of any size anywhere.
Are you available full-time?
No, this is outside capacity alongside a full-time platform engineering role. Good fit for a scoped project, an architecture review, or ongoing part-time/fractional work, not a full-time hire.
What does \"onboarding AI\" actually mean here?
Concretely: building custom Claude Code skills and agents for your engineering workflows, the same kind of automation used to turn a week of manual dashboard migration work into hours. Not generic AI strategy talk, actual tooling your team uses.
Do you work with stacks other than GCP?
Primary depth is GCP (BigQuery, Cloud Logging, Cloud Run, IAM) and AWS (EC2, Lambda, ACM), with Splunk, Grafana, Prometheus, and OpenTelemetry across both. Happy to discuss other stacks, just ask.