Write-ups of systems I’ve built that taught me something worth sharing. Each follows a consistent structure: the problem I was solving, constraints that shaped the solution, decisions I made and why, and what actually happened.
I include constraints explicitly because they’re often more instructive than the solution itself. A “good” architecture decision is only good in context—knowing that I chose X because of budget pressure or team size or existing infrastructure helps you decide whether the same choice makes sense for your situation.
A full on-prem infrastructure platform—from hyperconverged storage on commodity hardware through a 21-node Kubernetes cluster—with predictable costs, no cloud vendor lock-in, and rebuildable from scratch. Significant for deploying highly resilient infrastructure in air-gapped environments.
Integrating HubSpot CRM, Airtable, and a field-ready PWA so training coordinators have one place for operations—with live CRM data and reliable offline use.
