Wim Audenaert

Wim Audenaert did not set out to be a founder. He spent the first part of his career at Ghent University in Belgium, doing a PhD on the modelling of advanced oxidation (the ozonation and UV chemistry used to break down stubborn pollutants in water), and by his own account he came close to the professor track. The thing he kept coming back to was modelling, and in parallel he had what he calls entrepreneurial aspirations, so in 2017 he and a few colleagues spun a company out of the university. That company is AM-Team, and the bet behind it was simple: most of what an engineer learns from a physical pilot plant, they could learn faster from a good model.

Wim Audenaert's starting point is the slow, expensive way the water industry still develops new processes. When a treatment idea works at lab scale, the usual next step is to build a pilot, a real stainless-steel unit that someone has to operate, sample and tweak by hand, and that is where his frustration lives. He has watched teams spend a fortune building one of these and then, when it finally behaves, not actually understand why, which means they have learned nothing they can carry to full scale. His point is that the piece of that work which is about optimizing the design, as opposed to testing the chemistry in real water, should be done on a computer instead.

Wim Audenaert calls AM-Team's core service virtual piloting, which most of the industry now files under the buzzier label of digital twins. The engine underneath is CFD, computational fluid dynamics, the same physics-based simulation used to design aircraft and engines, here pointed at how water actually moves, mixes and reacts inside a reactor. AM-Team builds that detailed flow model, then has an in-house way to boil it down into a simpler process model that runs in standard simulators, so a client can keep using it. The promise he makes is that this does not take the months people assume, that his team finishes most projects in weeks, and that it can cut both the build cost and the running cost of a plant. AM-Team is deliberately not trying to replace the engineering and consulting firms that assemble treatment trains; it sells the optimization layer on top.

Wim Audenaert works across both drinking water and wastewater, and he is candid that physics-based models hit a wall on the messy side of that fence, because a wastewater stream is far more variable and harder to write equations for than clean drinking water. His vision is that mechanistic models and AI end up collaborating, with the AI filling the parts the physics cannot explain rather than replacing the physics. And he expects the craft itself to spread: the modelling that only specialists can do today, he thinks, will be something far more people do within a decade, the same way treatment-plant simulators have already become accessible to engineers who are not modellers. He sits on the International Water Association's Digital Water Programme steering committee, which is roughly the same argument made at industry level.

Wim Audenaert is, underneath the simulations, a scientist who got bitten by what he likes to call sciencepreneurship, the move from the lab into building a business. He is open that the hardest part of growing AM-Team was not the physics but learning to delegate, to take the tangle of jobs a founder does themselves and hand them, one by one, to someone you trust. It is a very human lesson from someone whose day job is teaching machines to model reality, and it is the kind of thing that comes out when you get him talking rather than presenting.

“I have literally seen people building a unit of like meters of height, stainless steel... hundreds of thousands of euros easily. And then it doesn't work. They manually open it, they start changing the design. Suddenly it works. They don't know why. So if you don't know why, you are not learning.”

Wim Audenaert is the rare deep-tech founder who is just as comfortable explaining a turbulence model as a cap table, and that combination is most of why a modelling spin-out from a Belgian university now sits behind the design of plants in more than twenty countries.