Hi, I'm
Engineering Leader · Independent Consultant · Runs His Own Cloud
17+ years building production systems at scale. Now independent: I move engineering teams off the hyperscaler onto infrastructure they own — phased, reversible, and costed before you commit. Cloud bills that grew faster than traffic are where every engagement starts.
I write about cloud costs, owned infrastructure, and the money side of running an independent tech practice.
My principle
Sovereignty stays a privilege of the sophisticated until someone absorbs the complexity and makes it ordinary. That's my work — making the tools that set you free simple enough to actually run.
Sovereignty and sophistication pull against each other — until the sophisticated becomes ordinary.
Every technology we've ever built has faced this problem, starting with fire. Fire, the hammer, the chisel — every tool carries its hazards and side effects, and that has always been the price of new technology. Anything that must be set up sovereignly demands know-how, and that is the real barrier we face today: most people simply haven't been shown how these things work. Before we push people toward more sovereignty, we have to teach them how to run what we're handing them.
I believe completely that sovereignty is the best path — and I also accept that today it is too technical, too sophisticated, for most people. That gap is not a reason to give up on sovereignty. It is the work itself.
This is my guiding principle, for my life and my business: I am here to resolve that complexity — to absorb it — so that ordinary people get the full benefit of sovereign tools without needing to master what's underneath.
Replaced $40K/month Datadog + $40K/month CloudWatch with a self-built Grafana LGTM stack. At 6 TB/day across 15 departments it ran ~$25K/month against a ~$225K SaaS projection — ~$200K/month of prospective spend avoided.
Architected a targeting engine with Java 21 virtual threads processing millions of events at <500ms p99.
Right-sizing, Karpenter, Graviton, and lifecycle policies across production infrastructure. Zero performance loss.
Run my own Kubernetes clusters, observability stack, and services on bare metal. No cloud lock-in.
Cloud costs, owned infrastructure, and the money side of an independent tech practice.
We switched Datadog off and saved $40K a month. Then a $40K/month NAT gateway charge appeared in EC2-Other, and nobody noticed for one to two months. The second-order cost of self-hosting your telemetry.
I run Talos — immutable, API-managed Kubernetes OS — on bare metal. The install, the no-SSH model, the upgrade scheme, and what's genuinely painful.
Soft multi-tenancy on Kubernetes has a ceiling, and it arrives sooner than most teams expect. What the rungs above it actually are — per-tenant network separation, dedicated VMs, child clusters — and what each one really costs to run.
I'm most active on Nostr and X. For longer conversations, email me.
Lightning: [email protected]
