About Herzog Enclave

Thirty years of making sure sensitive data ends up where it belongs — and nowhere else.

That sentence is the whole career. Everything else is detail.

Portrait of Larry Herzog Jr.
Larry Herzog Jr. · Founder and principal consultant

The throughline

It started in the U.S. Air Force, on tactical satellite and microwave communications — keeping critical information moving reliably and securely in conditions where failure wasn’t an option. That set the pattern I’ve followed ever since: not technology for its own sake, but protecting and moving information that actually matters.

From there came the decades of hands-on engineering that don’t make it onto conference slides — UNIX and Linux systems, enterprise networks, security operations. The infrastructure layer where things genuinely run and genuinely break. I’ve spent my career as an operator, not a spectator, and it shows in how I work.

Eventually that led into enterprise cybersecurity, and then to a deep specialization in its least glamorous, most consequential corner: data protection. Encryption. Key management. Hardware security modules. The unforgiving mechanics of making data useless to anyone who isn’t supposed to have it — and making sure the people who are supposed to have it never notice the difference. This is the core of what I do, and it’s where I’ve gone deepest.

On private AI, I’ve done the work — not just the talking

The newest frontier is running AI privately, and it’s crowded with people promising transformation who have never deployed anything.

I’m not one of them. I run a private-AI reference stack of my own, including large language models on local hardware, with sensitive data that never leaves the premises. So when I tell you that private, on-your-own-terms AI is achievable — that your people can have powerful tools without handing your data to a third party — it isn’t a theory I read about. It’s something I built, and something I use in my own work every day.

If you want to see how that experience turns into an actual client decision, review the illustrative Private AI Readiness Assessment. It shows the kind of architecture, risk judgment, and implementation sequencing the engagement produces without presenting an invented client story as proof.

The breadth is the offer; the depth is the reassurance

Along the way I’ve done most of what surrounds all of this: server and network administration across UNIX and Windows, physical security and surveillance systems, application development, inside- and outside-plant infrastructure.

That range is the point for most of the businesses I work with — one person who understands how the infrastructure, security, application, and physical layers interact, and where problems hide in the seams between them. The depth still matters when the work reaches data protection, encryption, or private AI. You should not have to choose between seeing the whole environment and understanding its most consequential parts.

Credentials, briefly

CISSP-certified. A master’s in Business Information Systems, finished at the top of my graduating class, with doctoral work in leadership set to begin. Decades of enterprise security experience behind all of it.

The letters matter less than the years — but you should know both are there.

How I work

No hype, in either direction. The AI world sells transformation; the security world sells fear by the pound. I’d rather tell you plainly what your real risk is, what’s worth doing about it, and what isn’t — then scope the work so you always know exactly what you’re getting and what it costs.

If the honest answer is that you need less than you feared, that’s the answer you’ll get. That’s rarer than it should be, and it’s the whole reason to work with one seasoned person instead of a machine that bills by the hour.

Based in Valley City, North Dakota. I work on-site with businesses across Eastern North Dakota and remotely wherever the work calls for it.

Book a working call