Your hardware, your data
Fadenstack runs where you run it. What leaves your network is a decision your IT makes, and the console shows it.
About
Practical AI shouldn't require an enterprise-sized team. Fadenstack is built by Emagin8 in Germany, for organizations that want AI for everyone but don't have whole departments to run it.
Why it exists
Bringing AI into an organization is rarely as simple as deploying a model or buying a chatbot. There are existing applications to connect, company knowledge to make accessible, sensitive data to protect, and infrastructure to keep running.
Large organizations put whole teams on this. Smaller companies usually can't. Fadenstack gives them one foundation for models, knowledge, tools and applications, inside the software they already use, so they can start small.
Who builds it
I'm an engineer and software developer. My work has taken me from production automation and industrial IT systems to software development and, more recently, LLM-based systems in everyday production use, from engineering tools to translation automation with measurable cost savings.
Those environments taught me that technology only creates value when it solves a real problem and fits the way people already work. Much of the engineering effort goes into connecting systems, making information accessible, and maintaining solutions long after their first deployment.
Building LLM-based systems, I kept running into the same challenges: every new application needed its own models, data access, integrations and infrastructure. That led me to a larger question:
What if companies didn't have to build a separate technical foundation for every AI application?
That question became LLM.Port, the open-source project that grew into Fadenstack. In 2026 I founded Emagin8 in Germany to turn it into one system that a small team can actually run.
What we build for
Fadenstack runs where you run it. What leaves your network is a decision your IT makes, and the console shows it.
Chat, agents, apps, models and machines share one set of users, rules and records, so IT runs one thing instead of a collection.
The server is open source under Apache-2.0. Anyone can read it, run it and keep running it.
Why the name
In German, der rote Faden (the red thread) is the guiding idea that runs through something and holds it together as a whole.
The expression goes back to Goethe's Die Wahlverwandtschaften (Elective Affinities, 1809). He describes a red thread woven through every rope of the English Royal Navy: it could not be pulled out without unravelling the rope, and even a small piece could still be recognized as belonging to the Crown.
Fadenstack is that thread through your organization's AI. Chat assistants, agents, data, models, tools and infrastructure are connected once, governed in one place, and traceable end to end.
Contact
Questions about Fadenstack, the planned Enterprise edition, or working together: write to us.