Hey, I’m Derrick — a Senior Cloud & Infrastructure Engineer based in Macomb, Michigan, with 18+ years spanning on-prem, hybrid, and cloud. My job, more often than not, is bringing order to complex environments: documenting the undocumented, right-sizing the over-built, and automating the parts nobody should be doing by hand.
Most recently I led a full datacenter-to-AWS transformation. I inherited an undocumented VMware environment — 15 hosts, a SAN, 100+ VMs, zero runbooks — audited every workload, established ownership, and decommissioned the dead weight. Then I migrated it service-first rather than lift-and-shift: native S3, AWS Backup, and CloudWatch in place of self-managed boxes, a right-sized SQL Server on EC2 where RDS didn’t pencil out, and a clean datacenter exit that retired all 15 hosts and the SAN.
I’d rather replace a server with a managed service than babysit it — and I’d rather understand every layer than rent a black box I can’t debug.
Alongside the cloud work I ran the day-to-day of a fully remote company: 800+ macOS and Windows endpoints managed and secured with Intune and Defender, 1Password rolled out company-wide to a 550+ person workforce, and acquisitions folded into a single Microsoft 365 tenant. I ran production releases to Kubernetes on AWS: tagging releases in GitHub to trigger image builds, deploying with GitHub Actions, and logging each release in JIRA. I also handled the cross-account networking, DNS, and database pieces of the EKS migration. I supported our SOC 2 compliance work, and I care about leaving documentation good enough that the next person doesn’t have to reverse-engineer anything.
How I work with AI
I use AI heavily, and on purpose. I treat it like a fast, tireless junior engineer: it drafts code, configs, and documentation; I set the architecture, review every change, and own what ships. The real skill isn’t prompting — it’s knowing a system well enough to tell when the AI is confidently wrong.
That approach is how I’ve built most of my recent side projects:
- Formpost — a self-hostable form backend for static sites, written in Go with SQLite and SMTP delivery.
- PushBridge — a self-hosted Apple Push Notification relay with a SwiftUI iOS app.
- RowMonitor — a SwiftUI/CoreBluetooth iOS app that talks to my rowing machine over BLE.
- Bucklio — an S3-compatible object storage server in Go (in progress).
- HavenSkye — a persistent virtual world I keep coming back to (an ongoing experiment).
At work I used the same approach to automate operational toil and build small internal tools. I’ll be writing about what works, what doesn’t, and where I don’t let AI anywhere near production.
When I’m off the clock
The curiosity doesn’t switch off at 5pm. My homelab is a handful of hosts running Docker, Traefik, and Komodo, tied together with Tailscale, with a strict one-service-per-directory layout and GitOps-style management. Alongside it I run Kubedoo, a Kubernetes cluster on Talos and Proxmox with ArgoCD, Cilium, Longhorn, and MetalLB. That’s where my deeper Kubernetes experience comes from.
Right now I’m using the lab to go deeper on the things I want more of: CI/CD pipelines, observability (metrics and centralized logging), and infrastructure as code. Expect write-ups as I go.
Linux has been the constant through all of it: at work, running email relays, print servers, websites, and authentication; at home, the entire lab. It’s the thing that taught me how much you can learn when you’re allowed to take the whole system apart.
Why this site exists
This is where I document what I’m building and running — cloud and infrastructure work, self-hosting, AI-assisted building, and the occasional hardware rabbit hole. If something took me a weekend to figure out, I’d rather the write-up save you an afternoon.
Thanks for stopping by. If you want to talk infrastructure, cloud, or self-hosting — or you’re hiring for a senior cloud or infrastructure role — reach out.