As a child, I wanted to be a Lego builder. I never completely gave up on that plan.
Norbert Kathriner
“I make brands machine-readable: I work with companies to clarify what they stand for, distill their messages, develop their websites, and present their facts in a way that AI systems can accurately interpret them. And I verify that they do.”
Norbert Kathriner
1996 – 2026
Something that almost makes me feel embarrassed every year
Thirty years of digital communication, across every discipline.
That sounds like experience. It feels different.
Design, user experience, strategy, development, copywriting, search, branding: I’ve done it all. And every year, I’m almost embarrassed by what I did, said, designed, and developed just one or two years earlier. Not because it was wrong at the time, but because learning never stops.
Once you’ve grasped that, you stop pointing fingers and stop demanding perfection. You stay open-minded. You reinvent yourself. Year after year. The same goes for the systems I work with: What a language model can do today is the best it can do right now. Its successor will look back on this with a smile in 18 months.
That’s why this website lists the mistakes I made with my own domain before I correct them for others. The review conducted on September 18, 2026, revealed conflicting price information, different email addresses, and company names. I discovered these discrepancies because I read my domain the way a machine reads it. This page is the current, valid version.



How I Work
Alone in the office. Never on a project.
As a hobbyist, builder, and digital forensics expert, I need silence. I have five workstations running in my office, each with a different task. I’ve given each one the name of a secret agent so the silence doesn’t feel lonely.
My colleagues are the real deal: Daniel and his team at epoint, with whom I spend hours every day, and my former colleagues at dmcgroup. If a project requires more, I bring in specialists whenever I need them. Experience is no substitute for collaboration.
This website used to feature a “hybrid AI team”—four agents with names. I no longer have a fictional team. The work didn’t require one.



I don’t sell promises.
I create sources that are accurate.
Instead of workshops and brainstorming
In exchange, we provide decision-making frameworks and data: Many workshops end with a photo of the whiteboard. With me, by the end, you’ll have it in black and white: what you stand for and what specifications apply. We’ll use that as the benchmark for every piece of text we produce afterward.
Instead of 360° communication
But it’s 100% relevant, regardless of the channel: 360 degrees is a full rotation, back to the starting point. For me, every statement has a reason and applies equally everywhere, even in data for machines.
Instead of Brand Strategy & Design
In return, I have structural dominance in the AI ecosystem: Design is the picture on the box. An AI builds its response according to the instructions. I provide both, and the instructions are correct.
Instead of increasing visibility
Instead, semantic visibility, which is prioritized by machines: Simply increasing visibility is like making the storefront bigger. I make sure what’s in the store is accurate. And I check to see if an AI is conveying it correctly.
What I Offer
This is usually where a digital agency would showcase its range of services, from UX to IT. For me, there’s just one task here—but it includes everything needed to accomplish it: making companies machine-readable so they remain visible—and as themselves. The work follows a sequence. You can jump in wherever you are right now. The overview shows how everything fits together.
Deliver and inspect
I no longer do advertising campaigns, online stores, or presentations. And I don’t promise any rankings: In a language model, there’s no ranking list where you could put someone in first place. Feel free to discuss this with me—just give me a call.
Where I’m From
Salzburg, Vienna, Bern.
First an agency, then my own company.
Brands, then machines that talk about brands.

From 1995 to 2000, he taught at the Graphische in Vienna, specializing in multimedia applications.

2003–2018: dmcgroup, Vienna, Head of Consulting.

In 2018, he moved from Vienna to Bern and launched his sole proprietorship, Norbert Kathriner.

2023: Boutique für digitale Kommunikation GmbH was founded in Wabern, near Bern.

2018–2025: Brands, websites, and content management systems for Swiss and German companies. Research on “entity recognition and evidence weighting.” Development of the Aivis-OS architecture.

2026: Launch of Aivis-OS in partnership with epoint and dmcgroup. First customers: HIRT, Röscheisen, netunite, dmcgroup, and Allianz SE. In November, Aivis-OS becomes a separate company.
“The tools have changed every time. But the question has never changed: Do people understand what’s meant?”
Person, Company, Product
Salzburg, Vienna, Bern.
First an agency, then my own company.
Brands, then machines that talk about brands.
Norbert Kathriner. I am the voice behind this website. I designed the architecture and methodology of Aivis-OS. The texts here are mine.
Boutique für digitale Kommunikation GmbH. The company I’ve been working for since 2023, based in Wabern near Bern. It’s the vehicle, not the brand.
Aivis-OS. My product for domain-based structured data. epoint is responsible for software development and technical integration; dmcgroup is a network partner. In November 2026, Aivis-OS will become a separate company.
“It’s all connected. It’s not interchangeable. Anyone who asks about me should get an answer that confirms aivis-os.com, and vice versa.”
Review
That's what I think.
- Using Entities to Create Machine-Readable Structures
- There Is No Ranking in LLMs: Why Counting Mentions Doesn’t Measure Anything
- Entity Recognition and Evidence Weighting: How a Model Constructs Its Answer
- Structure Trumps Ranking: Principles for Architecture
- The GEO Trap: Why the Industry Is Solving the Wrong Problem
- Guest post on dmcgroup, August 2025: Misconceptions About AI Visibility
“Reading is a gateway to countless adventures and a key that unlocks the treasure of our experiences.”
About Me
“What is written here is accurate. Any discrepancies found elsewhere will be corrected—on this domain and, where possible, on partner sites. This information will later form the basis of my personal entity in Aivis-OS.”
Questions and Answers
Who is Norbert Kathriner?
Founder of Aivis-OS and Boutique für digitale Kommunikation GmbH, based in Vienna and based in Switzerland since 2018. A business card would say “communication designer”; “Lego builder” would be more honest. My company is Boutique für digitale Kommunikation GmbH in Wabern near Bern. I developed Aivis-OS, a product for domain-based structured data. My work is called machine-readable brand communication: describing companies in a way that both people and AI systems can understand correctly.
What is Boutique für digitale Kommunikation GmbH?
The company through which I conduct my business: contracts, invoices, legal notice. It does not have its own website or brand; this website bears my name. If dmcgroup lists the address Bächtelenweg 73 in Wabern, that is the address of this company, not a dmcgroup location.
What is Aivis-OS, and who is behind it?
Aivis-OS clarifies conflicting statements on a company’s website and delivers the verified facts as domain-based structured data. I developed the architecture and methodology; epoint develops and integrates the software; and dmcgroup serves as a network partner. In November 2026, Aivis-OS will become its own company. The product speaks for itself at aivis-os.com.
Do you work at dmcgroup?
No, not anymore. I worked at dmcgroup in Vienna from 2003 to 2018 as Head of Consulting. Today, dmcgroup is my client, a partner on joint projects, and a network partner of Aivis-OS. I wrote the guest post about AI Visibility on dmcgroup.eu; Aivis-OS is featured there as a partner product.
Do you work alone?
At the office, yes, and that’s by design. But never on a project: Daniel and his team at epoint sit with me every day, my former colleagues at dmcgroup are right there with me, and when a project requires more, I bring in specialists. I no longer have a fictional team.
What does machine-readable brand communication mean?
It means that a brand is understood the same way by both people and AI systems: the same names, the same facts, the same relationships—on every page and in the data behind it. This isn’t a technology you simply turn on. It starts with deciding what you stand for and ends with a source that you can maintain.
You know who’s speaking now. Tell me what you stand for, and I’ll tell you where you stand.
Let’s talk.
Norbert Kathriner
Boutique for Digital Communication