Eindhoven startup Rigidbody, founded by three TU/e engineers, will pitch software that pairs physics-informed models with local language models to optimise mid-sized industrial plants at the AI Pitch Competition final on 12 November.
Credit: this report is based on a story first published by IO+. Read the original at ioplus.nl.
According to IO+, Rigidbody is one of the finalists of the AI Pitch Competition, whose final takes place on 12 November at the Evoluon in Eindhoven during AI Summit Brainport. The profile closes the outlet's series on the eight finalists and one wildcard.
The Eindhoven-based company is a spinout from a hydrogen initiative at Eindhoven University of Technology (TU/e). It was founded by Charlie Johnston (AI/ML lead), Rafael Marques Isakov (business strategist) and Alex Udrea (control systems engineer).
From student team to company
The founders spent a year building connections in the hydrogen industry through Splitter, an AI-based student initiative at TU/e. Splitter continues as a student team, while the three founders left to pursue Rigidbody.
That move meant walking away from an industry where they had strong connections, including a potential pilot partner, in favour of a more general AI product. The founders describe themselves as three students: a final-year computer science and engineering bachelor's student, a chemical engineer taking a master's in innovation management, and a mechanical engineer on a master's in mechatronics.
What the product does
Rigidbody combines physics-informed time-series models with local language models to build a plug-and-play intelligence layer for factory floors. The models aim to predict process performance, not just equipment failure, and the language layer is meant to make deployment possible without an in-house analytics team.
The software is not tied to a single equipment vendor. Inference runs on site by default so sensitive operational data can stay inside the plant, with cloud compute available when an operator needs it.
Market and next steps
One beachhead the team is exploring is the semiconductor supply chain, where many manufacturers are the mid-sized companies it targets. The founders say the Brainport region gives them direct access to that sector.
"The primary obstacle to scaling is establishing strong operator trust," the team told IO+. The founders also point to International Energy Agency figures that put industry at 24% of total final energy consumption in the Netherlands as of 2024. Their long-term goal is to become the standard intelligence layer for mid-sized manufacturers, starting in a single sector before scaling.
The AI Pitch Competition is organised by Erasmus Enterprise, BOM and IO+, with support from the Province of North Brabant.
Original source: IO+, "How Rigidbody brings plant intelligence to midsized manufacturing" Suggested image: Rigidbody team or product visual from the company website (rigidbody.nl)
