About
A researcher who went to run things
I lead digitalisation and IT at an industrial site. Before that I spent years studying why plants with plenty of data still decide by habit. The two jobs have more in common than either side admits.
Biography
I studied at TU Wien and spent the first part of my career in applied research. At Fraunhofer Austria I led the group for production optimisation and maintenance management, working with manufacturers in the automotive and semiconductor industries on how a plant can use what it already knows: in its machines, in its maintenance records and in the heads of its experienced people.
That work became a doctorate, completed at TU Wien in 2025, on knowledge-based maintenance: how to plan maintenance around the competence of the people who do it, and how to keep expert knowledge usable when the experts are busy, retired or on the other shift. It produced 18 publications, two of them in CIRP Annals, and a lasting suspicion of any system that works only in the demonstration.
Today I am Head of Digitalisation & IT at an industrial manufacturing site and a member of its management team. The job covers enterprise systems, production data and the practical use of AI. It comes with the constraints research tends to leave out: audits, safety, data protection and a plant that has to run on Monday. The projects that mattered most were less about software than about agreement: one forecast that sales and planning both trust, master data with a named owner, production losses measured the same way across the site, users who can help one another before they call IT.
I have learned to judge digital work by what changes outside IT. A system that goes live and changes no decision is a cost. So I start with the result, find its owner in the business and then choose the technology, in that order.
In my own time I run a small operating model of AI agents: four agents, one accountable person and an approval button on my phone. I treat it as a laboratory for management questions that industry is about to meet at scale. Who may decide what? What does cheap machine work really cost? How do such systems fail, and what must an organisation remember when its workers do not? I write about it in a series of essays, The firm of one, and I am careful about its limits. A firm of one is a wind tunnel. A factory is weather.
I teach at TU Wien as a guest lecturer, supervise theses and live in Austria.
How I lead
Six working rules
Digital transformation is rarely constrained by technology alone. The harder questions are where to invest, what to standardise, who owns the outcome and how much change an organisation can absorb.
Start with the result
A project begins with the operating number it should move and the person who answers for that number. The technology is chosen last, and is usually the easy part.
Find the owner outside IT
IT can build a forecast, a data model or a dashboard. It cannot own the decision they serve. Where no one in the business will put a name down, there is no project yet.
Standardise before you automate
Automating a process that three departments run three ways produces three fast processes and one slow argument. Agree first.
A number needs a decision
A metric with no owner and no decision attached decays, however handsome the cockpit. For each one I ask what it changes and who acts on it, and delete the rest.
Treat failures like stoppages
Systems break at handovers and through defaults nobody chose. A failed job gets the same review as a stopped line: what happened, who noticed, what changes. Blame is not on the form.
Build capabilities, not projects
A go-live is a date. What counts is whether the site can do something a year later that it could not do before, without the project team in the room.
Route
Research, applied technology, management
- 2019
Diploma thesis, TU Wien
On automatic recommendations for prescriptive maintenance. The start of a long interest in what machines should suggest and people should decide.
- Then
Applied research, Fraunhofer Austria
Industrial AI from the laboratory to the production floor, later as group leader for production optimisation and maintenance management. Projects with automotive and semiconductor manufacturers.
- 2021 and 2023
Two papers in CIRP Annals
On text mining for failure detection in production systems, and on planning maintenance around people’s competence.
- 2025
Doctorate, TU Wien
Dissertation on a knowledge-based maintenance framework for smart manufacturing.
- Today
Head of Digitalisation & IT
At an industrial manufacturing site, as a member of the site management team.
- 2026
The firm of one
A private operating model of four AI agents becomes a management case, and from December a series of essays.