I spent 20 years in the United States Air Force. Enlisted to officer. E-1 through E-7, then O-1 through O-3. Most people don't get the kind of career the military gave me — not the rank, but the exposure. Very few paths put you across that many domains, with that many brilliant people, in that short a time.

I started at Barksdale AFB working on cruise missile telemetry — signals processing, instrumentation payloads, SIGINT collection, and test systems that don't leave room for error. This was weapons-grade data infrastructure. Your code either produces clean telemetry or someone makes a very expensive mistake. That's where I learned to think about systems as things that have to work when nobody's watching.

From there I worked communications across three continents — satellite links, tactical radios, nuclear command and control systems, the kind of infrastructure that has to survive when everything around it is designed to kill it. Milstar terminals, HEMP-hardened shelters, encrypted ground-to-air links. I got deep into electronic warfare and ISR — SIGINT collection, emissions control, the intersection of RF engineering and operational security. I supported operations across EUCOM, AFRICOM, CENTCOM, NATO, and 52 coalition nations. I learned how systems detect each other, deny each other, and survive each other.

As an E-6, I took over base infrastructure — 470 miles of cable, fiber, and antenna systems supporting five combatant commands and 52 coalition nations. That wasn't even my career field. I learned networking from the ground up, passed our inspection with the highest score the infrastructure shop had ever received, and my team won an AMC MAJCOM-level award. I earned a 3.91 GPA finishing my degree at Indiana State while doing all of it. Not because I'm academic — because I don't do things halfway.

Then I commissioned and went cyber.

Network defense. Threat hunting. Incident response. Vulnerability assessment. Forensic analysis. I managed every layer of the defensive stack — firewalls, IDS/IPS, SIEM platforms, packet capture, log correlation, endpoint detection. Not as a consultant. Not as an auditor. As the person responsible when something got through. Distinguished Graduate from Undergraduate Cyber Training.

At AFCERT, I ran a flight of 160+ operators defending the Air Force Information Network — one of the largest enterprise networks on the planet. We closed over 4,000 investigations and eliminated critical intrusions across multiple combatant commands. That flight is now run by three captains. I was the lowest-ranking company grade officer in the squadron, but I was a prior enlisted master sergeant, so when the hardest job in the unit opened up, my commander handed it to me. 160+ operators, dozens of specialties, real-time threat response across a global network. I learned more from them than any school I've attended.

I was on a team that built a network attack simulator — a self-contained training environment deployed across multiple wings for offensive and defensive cyber exercises. You had to model real adversary behavior, build the infrastructure to contain it, instrument every layer for observation, and make the whole thing portable enough to ship to the next base. That project is the closest thing in my military career to what I'm building now.

My last assignment brought it full circle. Twenty years later, I was the Executive Officer for the Air Force's largest training wing. That's where the Air Force opened up AI tools for official use, and that's where I started learning what they could actually do. I deployed AI to eliminate 79% of unnecessary reports, reduce holdover students by 76%, and save $30M annually. I saw what AI could do inside a military wing. Now I'm building it for everyone.

The same thing I've always built. Secure infrastructure. The difference is who it's for.

Networks. Firewalls. VLANs. DNS. Routing. Segmentation. Redundancy. Failover. Distributed systems that keep running when parts of them don't. I learned this on the enlisted side maintaining infrastructure, then again on the officer side defending it. Now I build it for the people who never had access to that level of engineering.

Proxmox for virtualization. OPNsense for firewalls and network segmentation. Linux everywhere. Containers where they make sense, bare metal where they don't. Layered defense, zero trust between segments, everything monitored, nothing phones home unless I tell it to.

I taught myself AI/ML deployment from scratch. I run models from 1B to 120B parameters on local hardware — AMD, Nvidia, Apple Silicon — with inference speeds that compete with cloud APIs. No remote servers. No API calls to someone else's infrastructure. Local inference, local data, local control. I'm not a data scientist. I'm an infrastructure engineer who figured out how to make large language models run on hardware that fits on a desk — and then built the systems around them to make that actually useful.

I filed a patent for a mechanism I call the Differential Inference Engine — a validation layer that forces every AI decision to be contested before it's trusted. I designed the full platform stack: a local runtime for inference across heterogeneous hardware, timeline-based data collection through beacon nodes, and application-layer lenses that shape raw data into whatever the user actually needs — smart home automation, business operations, network defense, autonomous agents.

If you don't own the hardware, you don't own the intelligence.

I used to lead 160+ operators defending a global network. Now I lead a team of five AI agents on local hardware. Same principles. Different kind of team. The agents monitor the network, manage the household, handle business operations, inspect each other's work, and run security checks. They coordinate, they escalate, they report. And they do it whether the internet is up or not.

I used to build SOCs for the government. Now I'm building one for your living room.

Twenty years of building secure, resilient infrastructure for the government. SIGINT systems, cyber defense, electronic warfare, nuclear command and control, networks that survive when everything around them fails. That's the foundation.

Now I'm building it for everyone else.

The technology stays local. The data stays yours. The AI runs on your hardware, on your network, on your terms. When the internet goes down, nothing changes. When the power grid fails, a solar panel and a battery keep you running. When a model gives you a bad answer, the validation layer catches it before you act on it.

For the homesteads running solar. The towns building mesh networks. The families who want a smart home that actually works offline. The small businesses that can't afford a SOC but still deserve one.

Don't rent your intelligence. Own it.