Sam Rominger hits eight years at Koltiv this January.
His title is Integrated Technologies Engineer, Storage and Virtualization. Ask him to describe a normal day in one sentence, and you get three key areas: writing the scopes of work for projects before they start, moving workloads to the cloud, and supporting client infrastructure once it gets there.
He came to Koltiv from Microsoft, where he worked in the data center stack inside their facilities around Des Moines. Hardware upgrades, replacing parts, running cable, decommissioning gear, and supporting the Azure workloads sitting on top of it all. So when Sam talks about what runs underneath the cloud, he is talking about racks he has physically had his hands on.
He started at Koltiv in the NOC and landed in engineering two and a half years later.
Caffeine of choice? Coffee. Three packets of sugar and a little creamer. ☕
Morning person or night owl? Night owl. Up until two or three in the morning. 🌅
Show you have watched too many times? Family Guy. It just runs in the background while he is up, and has since he was a kid.
Something you are weirdly good at that has nothing to do with your job? Cooking. Ribeyes on the stovetop. Pork chops breaded and seared, then baked about twenty-five minutes.
We asked Sam what part of his job you would find interesting if you could watch it happen.
"Probably all the back-end support that I do in the cloud." Maintenance he handles and then tells you about afterward. Once your workloads are off your own hardware, most of you never see the environment again.
"All they know is it just works."
The project he is proudest of lately is the best example of that. The client barely registered it happening. The team replicated everything ahead of time and cut over the night before. People logged in the next morning to an environment that was simply faster, with the file server responding quicker than it had on their own equipment. The reaction, as Sam tells it: why didn't we do this a long time ago?
If you only ever knew Sam existed for one reason, he would want it to be that. Not the cutover itself, but what comes after. An environment with real fault tolerance behind it instead of a single box in a closet.
Sam has been deep in Microsoft Copilot lately, so we asked him to explain it to someone who has heard the name and nothing else.
"I explain it more like it's a personal assistant. Everyone gets a little personal assistance." Research, drafting, and finding the thing you know you made and cannot locate. He uses it on his own OneDrive when somebody asks about a process he wrote up years ago, and he has no idea where it went.
Then we asked what usually gets skipped when a company wants to switch it on for everybody at once. His answer was not licensing or training. It was data loss prevention.
Most businesses are not using DLP today, and plenty have never had anyone bring it up. That was survivable before. It is a different question once AI can query anything an employee has access to, because Copilot will surface files people forgot were shared.
His example is the ordinary kind, not the dramatic kind. Someone emails a copy of a Social Security card to HR and hits an all-staff or external list by mistake. Internally, you have options for pulling that back. Externally, it is gone. A policy set up ahead of time would have caught it before it left.
None of which is a reason to sit out AI. It is a reason to look at your own data first.
His advice for keeping your standards high while things speed up is the same thing he tells our team: trust but verify. Check the output against what you actually asked for, every time, until it earns the trust.
Sam is one of the people building our Copilot Enablement Program. If the data loss prevention piece above made you wonder about your own environment, that is the conversation worth having first.