Nuclicore

Nuclicore is a compliant enterprise no-code platform tailored to financial institutions including insurance, banking, asset management, and capital markets.

Founder

Covomo Versicherungsvergleich GmbH

Co-Founder & Managing Director

HeavenHR

Director of Insurance

EBS Universität für Wirtschaft und Recht

Postdoctoral Researcher/Doctoral Candidate/Research Assistant

Freelance Management Consultant

Management Consultant

CTG Corporate Transformation Group

Associate

I haven't posted here in a while. The last few months went into something more important for me: long conversations with customers about how they actually build software.

"We do not need more software. We need better software."

This is what a managing director told me and the line stayed with me.

Many companies ship new tools all the time. New frontends, new dashboards, new portals. Under the surface you often see the same pattern. A shiny UI on top of a backend that nobody really wants to touch anymore. After a few years the frontend still moves, the core feels frozen.

In most cases the problem is not effort or talent. It is how the work starts. Teams jump straight into screens and tickets before the basics are really clear. Who should see what. Which data flows matter. Where things can safely fail. How different systems talk to each other. All of that is added later, and every "later" makes the system harder to change.

When I talk to experienced architects they already think in roles, domains, contracts, non functional requirements and blast radius. The strange thing is how rarely this thinking is captured up front in a way that tools and teams can actually follow.

AI will not fix this by generating even more code. The real opportunity is different.

Start from the outcome. Ask the hard questions about roles, data, risks and success. Turn the answers into explicit tasks with clear acceptance criteria. Then let systems and agents build and test in small, reviewable steps.

If we do that, app builders and AI tools stop being "feature factories" and become assistants for designing software that stays healthy over time. Frontend and backend evolve together. Governance and speed are not enemies anymore.

This is the direction we are working on with Nuclicore right now. Not just more software, but better software.

How do you make sure the software you ship is "better", not just "more"?

Dr. Eberhard Riesenkampff

Founder at Nuclicore | Build production software with AI

Stop building MVPs, it's a waste of your time and money.

Managing Director told me a story that I hear more and more from different industries. 
They tasked their tech team with quickly testing an idea. Build a small MVP, nothing fancy. Two sprints, a few screens, a simple backend, works on demo day. Everyone claps.

The MVP worked. They wanted to go real with it. That is where the pain started.

Nothing from the MVP could be reused.
Not the data model.
Not the logic.
Not the integrations.
Not even the UI structure because it was stitched together for speed.

So they started again.
Version two.
More teams involved, new budget.
This time "for real".

Costs and timelines exploded. Even if compared to their original planning for the real thing. No one was happy.

The MD said to me: "We thought we were validating a business idea. Instead we validated how fast we could burn money."

I hear versions of this a lot. We pretend MVPs are cheap experiments. In reality they often create tech debt before a real project even starts. Once people see something that kind of works, the pressure to turn the toy into business software is huge. And the rewrite usually hurts more than anyone expected.

Don't get me wrong, getting sth out there to test is great and you should continue to do that. But build the foundations right so that they can be used for the real product. Or build a click-dummy and don't call it MVP 😂 

Curious how you handle this. Do you still build classic MVPs, or did you change your approach after a painful rewrite?

Nuclicore

Gestern Kickoff im Venture Clienting Programm von ekipa 🧠⚡

Viele Menschen im Raum, sehr viele Slides. Und zum Glück auch ein paar richtig spannende Corporates: zum Beispiel Caritas, EAM | Energie aus der Mitte, MTU Aero Engines und TRUMPF und ein paar weitere Schwergewichte aus Energie, Finanzwelt und Industrie.

Die Grundidee gefällt mir: statt wieder einen „Innovation Award“ an die Wand zu hängen, entstehen echte Pilotprojekte, die im Kerngeschäft landen und ausgerollt werden und nicht im Lab verstauben.

Wir sind mit Nuclicore dabei, der AI Builder Plattform, mit der Unternehmen aus kurzen Beschreibung echte Business Software planen, bauen, testen und auf deployen. Software, APIs und Automations, die auch neben den alten Bestandssystemen sauber laufen. Alles Enterprise-ready und compliant. Klingt nach Win-Win für mich.

Bin gespannt auf die folgenden Gespräche.

A CIO recently told me "Eberhard, shipping new software is not my problem. Making it work with what we already have is."

This is what I'm hearing a lot recently. 

There are rarely greenfield situations. In reality, almost everything that matters lives in brownfield. Core systems, half documented APIs, brittle scripts that nobody wants to touch, data scattered across five tools.

New portal looks great. Old policy system stays weird. Finance has its own truth. Ops has another one. Every new app adds a bit of value and a bit of complexity. Over time, the complexity wins.

The interesting part for me is that AI can either make this worse or much better.

Worse if it just generates more isolated apps and code.

Better if you treat it like a team of AI agents that understand the outcome you want, map the existing systems, propose contracts, glue code and migrations, and then build and test changes so they fit into the landscape without breaking everything else.

That is what is important to our customers and also to us. With Nuclicore, there are less new toys on the side, more "this has to live next to the old stuff without creating another mess."

Curious how you handle this today. When you ship something new, how much of your energy goes into integration with what is already there?

taplio

Dr. Eberhard Riesenkampff's Linkedin Analytics

Dr. Eberhard Riesenkampff

Performance ScoreDr. Eberhard Riesenkampff's LinkedIn Performance

Dr. Eberhard Riesenkampff's Best Posts (last 30 days)

Ready to grow your LinkedIn presence?

taplio

.css-6ytdur{background-image:linear-gradient(284deg, rgba(39, 43, 72, 1) 0%, rgba(33, 47, 183, 1) 100%);color:transparent;-webkit-background-clip:text;background-clip:text;}Dr. Eberhard Riesenkampff's Linkedin Analytics

Dr. Eberhard Riesenkampff

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Dr. Eberhard Riesenkampff's LinkedIn Performance

Dr. Eberhard Riesenkampff's Best Posts .css-1emoo84{font-size:var(--chakra-fontSizes-md);font-weight:400;color:var(--chakra-colors-gray-600);}(last 30 days)

Ready to grow your LinkedIn presence?

Dr. Eberhard Riesenkampff's Linkedin Analytics

Performance ScoreDr. Eberhard Riesenkampff's LinkedIn Performance

Dr. Eberhard Riesenkampff's Best Posts (last 30 days)