Javier Pérez

The first idea for RAGnexus came to me three years ago, at a talk on RAG technology at the Mobile World Congress. I had never heard of the technology before. But the moment I understood what it was for, I knew that AI personalized to a company's knowledge and needs was the future.

Later, as other companies pushed AI forward, I realized that the ultimate goal for getting the most out of AI is having it fully connected to your company, with agents that act like your own employees.

That's where the brand slogan comes from: Frictionless AI Universality.

This also carries serious security implications, which is why it matters so much to choose carefully which provider you trust.

When you paste a contract into an AI chat, that contract leaves your company. It lands with a provider you don't control, under terms almost nobody reads. And even if the server sits in Frankfurt, if the company running it answers to another country's laws, so does your data.

Meanwhile, people use AI anyway. Quietly, because it saves them hours. And companies are left with an absurd choice: give up on AI, or hand someone else the most valuable thing they have, which is what they know.

That choice shouldn't exist. That's why I started RAGnexus.

What a company knows

Every company knows far more than it thinks. That knowledge lives in contracts, email threads, shared folders, Slack channels, and the head of whoever has been there fifteen years. The problem isn't a lack of information. It's that nobody can find it.

Every new tool promises to fix this and adds another tab. One assistant in your inbox, another in your documents, another in your CRM, each with its own subscription, its own rules, and its own access to your data. AI was supposed to bring it all together. Instead, it's as scattered as everything else.

What I set out to build

I wanted something easy to explain and hard to build: one AI for the whole company, living where the company's data lives.

An AI connected to the tools you already use (SharePoint, Google Drive, Notion, Slack, Salesforce, and more than forty in all) that looks in your own files first when you ask it something, and shows you which document the answer came from.

Personalized in the way that counts: it speaks your company's language, knows your clients, and understands how your team works. Without training anything on your data: it doesn't need to, it looks things up every time you ask.

All-in-one AI, in the literal sense. Not another tool, but a workspace: the place where your team asks, writes, analyzes, and creates documents, spreadsheets, and presentations without jumping between apps. An ecosystem instead of a pile of tabs. The tools you already have, connected. The ones you're missing, built in.

And it should explain itself. I gave myself one rule: if anyone needs a training course to use it, I've failed.

Why Europe

I could build this anywhere. I'm building it in Europe, on European servers, and it isn't about waving a flag. The infrastructure is European because the legal framework is.

Digital sovereignty isn't about having the server nearby. It's about knowing who controls it, which laws it runs under, and who it answers to.

What the General Data Protection Regulation (GDPR) calls data protection by design, in Article 25, isn't a box I tick at the end. It's where I start.

That's why RAGnexus runs on European servers or, if you need it, on your own. You choose your level of sovereignty. It's also why I'm not asking you to take my word for anything. Every answer comes with its source, so you can check it in one click before you rely on it.

Europe has taken privacy seriously. What makes no sense is that its companies have to check it at the door to use AI.

Where it stands today

RAGnexus is young, and I won't pretend otherwise. It already connects to more than forty tools, cites its sources, and goes live in about four weeks with no technical team needed on your side. And it's starting to create documents, spreadsheets, and presentations without leaving the platform.

What comes next is an AI that doesn't just answer but also takes on the repetitive work nobody will miss, so your team can spend its time on what actually takes judgment.

Javier Pérez, founder of RAGnexus

Frictionless AI Universality — private AI grounded in your company's knowledge.