cCortex genesis

»Everything should be made as simple as possible, but not simpler.« Einstein

How an integration problem turned into a new system principle

For almost a decade, I worked in the Group COO Office of a major financial institution.
During the setup of new subsidiaries, we faced a persistent challenge:
their processes and control systems simply didn’t fit into our standardized ERP world.
New business models demanded far greater flexibility — something our established core systems couldn’t provide.

When standards fail, you fall back on custom solutions.
But custom systems inevitably lead to compatibility and integration headaches.
Since we had to build something new from scratch anyway, I decided to go one step further:
to design a solution so fundamental that all future integration problems would disappear by design.

The smallest common denominator

What is the smallest common denominator of all control and calculation systems?
A network of interdependent variables.

But real networks are really complex.
Their structures are non-hierarchical, and traditional algorithms struggle to handle them efficiently.
Recursion and dependency chains cause exponential growth in complexity, runtime issues, and memory limits.
I needed a way to perform all those calculations without recursion —
to make the network solve “itself”.

The idea of bubbleCalc

The breakthrough came from a simple metaphor.
What if every variable could carry its own context — like a small traveler with a backpack —
and go on a journey to resolve “itself”?

Each variable would collect whatever intermediate results it needed,
and once all conditions are fulfilled, it would signal its completion to the network.
Other variables could then use it for their own resolution —
just like bubbles in a glass of champagne that rise, connect, and lift each other upward.

From this idea, I developed bubbleCalc,
a decentralized, recursion-free calculation process,
and a corresponding coordination layer called bubbleHub.

Unlike the classic bubbleSort that inspired its name,
bubbleCalc is not heuristic but a high-performance algorithm.
It avoids the pitfalls of recursion and performs with remarkable efficiency,
regardless of scale or structural depth.

From concept to patent

Because the approach worked exceptionally well,
I filed an official invention disclosure under German employee-inventor law.
Algorithms themselves are not patentable,
so we protected the underlying functional data architecture as a computer-implemented invention (CII).

After a successful proof of concept —
and a series of internal restructurings —
the rights to the invention were ultimately transferred to me.
It has since been filed internationally,
with the first patents already granted and others pending with positive outlooks.

Where things stand today

The original problem — seamless integration of complex control systems —
has evolved into something much larger:
a universal, lossless system architecture that unifies control, data, and computation in one dynamic model.
The cCortex framework now scales effortlessly,
handling anything from enterprise integration to truly dynamic neural networks —
which, at their core, are also just variable networks.

Today, I use this website — still a bit of a personal work in progress —
to share my ideas,
to connect with developers, researchers, and partners
who are as excited about scalable simplicity as I am.

Thank you for taking the time to explore it —
and for your understanding that what you see here is still evolving,
much like the system it describes.