About
My parents bought me my first computer for Christmas of 1991 — I was in ninth grade. A 386 DX40, 8 megabytes of RAM, a 120-megabyte hard drive, an external SCSI CD-ROM, a Sound Blaster Pro, and a dot-matrix printer that woke the whole house every time it ran. Over the next six years that machine kept changing — a grayscale handheld scanner, a color printer, and a steady drip of motherboard, processor, memory, and storage upgrades, every dollar of birthday money turned into the next bus speed. By the time I finally replaced it I'd already learned the most important thing a computer could teach you: that the substrate is fixable, hackable, and yours.
I've spent the thirty-some years since chasing the sound of those machines. The hum of platters. The smell of warm capacitors. The patience of waiting for something to compile.
The world has gotten better at hiding all that — buried it under enough layers of abstraction that most people don't think about what's actually running anymore — but the shape of how we got here is still in everything we touch. Every cloud region. Every container. Every model weight. Echoes of the Machine is what got left behind, and what's about to become the next echo.
Who I am
I'm Sid Smith. Solutions engineer by trade, technologist by gravity, homelab nerd by choice.
When I'm not at a keyboard I'm usually outside. Paddleboarding, hiking, biking, camping, scuba diving (mostly Florida, willing to travel for a good wreck), hunting, and the kind of just-go exploring that doesn't have an itinerary. Photography is its own thread — I picked up a diploma from the New York Institute of Photography along the way — and the drones started as an extension of that, partly for the aerial work and partly because I just like seeing places from a different angle. I'd rather be on a trail, a boat, or a board than almost anywhere else.
The thing I keep coming back to — and the lens I use for almost everything I write here — is that balance is what keeps us human while we adopt the tech that's supposed to make our lives better. Technology built for its own sake, or for nothing more than turning a profit, misses the mark. The only honest metric is whether it improves the quality of human life. If you're stuck behind a screen fourteen hours a day, you're not doing it right. We weren't built for that.
The career version: I started in K–6 IT, then took on a full data center and network transformation at Pierce College — virtualized roughly 80% of the college's workloads, brought in a Fibre Channel SAN, stood up virtual-to-physical clusters and full disaster recovery. That work made me an early adopter of virtualization, before "cloud" was a word people threw around, and I spent the next several years repeating the same playbook for other companies as a virtualization consultant. Around that same era I co-founded an ISP that put broadband into school districts the big carriers wouldn't touch. And then an early-2000s detour I wear as a badge of honor: I founded a company that helped bulk emailers — yes, spammers — get their mail through when the CAN-SPAM Act first dropped. Spamhaus listed me by name as a safe haven for spammers. Every chapter teaches something. The consulting work eventually put me in the room when private cloud was being designed for the first time: early SE at DynamicOps, then on to VMware after they acquired us. The platform we built there became vRealize Automation, and lives on today inside the VMware Aria stack. While I was VMware's National Cloud Specialist I also worked on the early design, testing, and implementation of Codestream — VMware's CI/CD platform — which led into the last ten years of work in IaC, Kubernetes, Cloud Native, and AI/ML. Founding-team SE at SovLabs, Office of the CTO at CloudBolt, first SE at Firefly.ai. The pattern's been consistent: show up where the next era is taking shape, and try to make the transition real for people who aren't already at the front of the line.
These days that means working on the same throughline I picked up running that K–12 ISP twenty-some years ago: how to put real, owned, locally-hosted technology into the hands of people who get nothing useful from a hyperscaler's roadmap. Local communities. Small businesses. Individuals. Locally-hosted AI. Real ownership. No recurring cloud subscription. The hyperscalers don't sell to small towns. Someone has to.
I also run my own personal cloud at home — a cluster I call Helix that hosts most of the services I rely on day to day, including the pipeline behind this blog. Owning your infrastructure changes how you think about everyone else's. That tension shows up a lot in what I write.
What this blog is
I started writing again because for the first time in a long time I have hope.
The last decade of consumer technology — social media, in-app purchases, subscriptions for everything — did not turn out to be the leap forward we were promised. SaaS, in my view, is dead. It helped vault us through the cloud era, and it also taught a generation of builders that the correct business model is to extract a recurring monthly tax for software that used to be a thing you owned. People are tired of the technology tax. I'm tired of the technology tax.
Then AI showed up, and the obvious next move from the same playbook was to charge a few hundred dollars a month for the privilege of using it. That isn't a utility model. That's a toll booth in front of a public road.
The thing that gives me hope is that the hardware finally caught up. The model weights are out there. The local inference stack works. A capable AI assistant is now something you can own rather than something you have to rent. For the first time in a long time, the door to opting out of the technology tax is open — and most of what I'll be writing here is documentation of me walking through it. Winding down the SaaS subscriptions. Pulling my data back to where I can see it. Rebuilding what I need from the ground up, with AI doing some of the heavy lifting.
The site also carries the dailyhypervisor archive — pieces I wrote between 2009 and 2016 about helping companies build the early private, public, and hybrid global cloud architectures that defined that decade. vRA, broader cloud automation, orchestration, and the work of turning racks of bare metal into self-service stacks across the platforms of that era. The technology has aged; the patterns of thinking about it haven't. Same writer, twenty-some years apart, working through the same shape of problem in different decades.
Recurring threads
A few arcs I keep returning to. If you want a way in:
The Imprint arc — a long-running thread on personal AI ownership: who owns the artifact when a model is shaped by a person's expertise, what a portable expertise object would actually look like, and where the supply-side and licensing infrastructure still has to be invented. It started in 2023 with a thought experiment, and the 2025 revisit tracks what changed once the substrate caught up. The whole framing piece hangs off it.
Lived-in agentic operations — what working with coding agents every day actually teaches you. Plan-mode discipline, blast-radius thinking, audit trails, the metaphors that turn out to be useful. The canonical entry is the day an AI agent deleted production. The official documentation does not cover this material.
Cloud and automation — still the bones of everything, still worth understanding deeply. The category I built my career on doesn't go away because there's a new category on top of it. Most of the dailyhypervisor archive lives here.
Local-first AI — the slow shift from AI-as-a-service-you-rent to AI-as-a-thing-you-own. Increasingly Apple-Silicon-shaped on my own desk; increasingly the question of how this stack reaches the people the hyperscalers ignore. Not as the vendors want me to talk about it.
Looking back — periodic pieces that connect the current moment to what came before. The way the field temporarily abandons fundamentals. The way the loud voices win the early innings and the patient voices win the long game. The way the ground rules of ownership and licensing get rewritten in fights that the engineers building the systems don't see coming.
What you'll get here
Personal posts. Technical writeups. Opinionated takes when I have one. Not press releases. Not vendor-friendly. Not "five tips" listicles. Mostly: what I've actually built or learned, written in the language I'd use with another engineer over a beer.
The cadence is roughly two posts a week, more when something demands it. The pieces will sometimes be wrong, and when they are wrong I'll come back and write the second piece that says so.
If you find something useful, let me know. If you want to argue, even better.
— Sid