Scaling Robots with Purpose | Robert MacKenzie, Ellipsis Venture
Robert MacKenzie, Venture Partner at Ellipsis Venture and Head of Product Management at TWINT, on structured commercialisation, product leadership and physical AI hype.
Discuss this episode and meet other Swiss tech builders in the TechDrive WhatsApp group.
Key insights: executive summary
- Premature commercialisation can push unfinished deep tech products into the market before the company is operationally ready.
- Robotics deployments should be structured as strict learning loops that feed real-world evidence back into product requirements.
- Academic founders risk losing their scientific strength when they stop testing hypotheses and become attached to their first solution.
- The humanoid and physical AI market may face an investment correction as expectations move ahead of deployable results.
- Hardware companies benefit from independent quality assurance and test engineers who are willing to challenge the founding team's assumptions.
“We're now 400 workers, and 200 of them are human.”
“Calling your baby ugly is not easy, especially for founders.”
What is structured commercialisation and why does it matter now?
Robotics and deep tech cannot be commercialised like software as a service. Sending an expensive robot into the field without explicit learning goals creates costly activity without reliable progress. Structured commercialisation simplifies the deployment, defines a learning agenda and acknowledges the complexity still hidden in the hardware system.
How can deep tech spin-offs cross the chasm to real products?
Build teams that balance visionary engineering with operational friction. Test and quality-assurance specialists should challenge designs early and validate them against the disorder of real customer environments.
Do not lose your scientific advantage
Keep the academic learning loop alive. Test hypotheses continuously and allow the solution to evolve with real-world evidence.
Start with the customer's reality
Even when the technology is ready, the customer's environment and data-sharing policies may not be. Work backwards from adoption.
Who is Robert MacKenzie?
Robert MacKenzie is a Venture Partner at Ellipsis Venture and Head of Product Management at TWINT. His earlier work at ANYbotics, Labster and Hocoma gives him a practical view of moving Swiss deep tech from research into reliable products.
Episode show notes
Robert MacKenzie helps technology teams make the transition from the lab to deployable products. Drawing on experience scaling companies including ANYbotics and Labster, he examines the difficult work of turning deep tech into reliable market solutions.
What stood out was his direct view of the physical AI hype cycle. AI is a powerful enabler, but the physical world remains the bottleneck. The companies that endure will structure commercial deployments as rigorous learning loops.
More conversations from the Swiss tech ecosystem
- Nobody Has Left a Robot Working Overnight | Yulia Sandamirskaya, ZHAWYulia Sandamirskaya, ZHAW Zurich University of Applied Sciences
- Physical AI Is Ready, Buildings Are Not | Kateryna Portmann, ANYboticsKateryna Portmann, ANYbotics
- Switzerland's AI Movement Is About to Start | Swiss {ai} WeeksKatka Letzing and Anna Hakobyan, Swiss {ai} Weeks
Follow the Swiss tech ecosystem, one drive at a time
New in-car interviews with founders, investors and researchers building Zurich's deep tech, AI and robotics scene. Want to discuss this episode, ask follow-up questions and meet other builders?