I want to explain why I find this contract more interesting than its size. I think it says something about where this industry actually is.

There is a nice piece of history here, and I think it is relevant. The first working programmable digital computers in the world were built in Germany in the early 1940s by Konrad Zuse, and he built them for a very unglamorous reason: Aerodynamics engineers were overwhelmed by repetitive structural and aerodynamic calculations, including wing-flutter analysis. Eighty years later, Germany’s aerospace research center is procuring quantum computers, and I notice the spirit is the same. Machines acquired to do work, not to make a point.

DLR QCI is not a company kicking the tires. It is one of Europe’s most serious quantum research programs, funded by the German Federal Ministry of Research, Technology and Space, and over the years it has commissioned quantum computers across nearly every technology there is. These are people who know exactly what they are looking at. When a buyer like that adds superconducting systems to its portfolio, it is not a bet on a promise. It is a judgment about what runs in production today.

Now look at what they asked for. Nowhere in the requirements will you find a race for qubit counts. What you find instead are practical questions. Can their teams manage their own users and accounts? Can researchers take an algorithm from first plan to finished run on one platform? Can they benchmark the new systems against the machines they already operate? As a physicist, I will admit that the PhD-student version of me would might have found this list boring. Today I read it as the most encouraging document in the deal. And because real research problems do not politely split themselves into quantum work and classical work, the contract includes a dedicated classical server in our Munich data center, sitting right next to the quantum systems, so DLR QCI can run workloads that combine both.

I have written before that a technology grows up when its conversations become boring. This contract is full of gloriously boring things. Account administration. Benchmarking. Integration. Zuse would have recognized every line of it. That is what procurement looks like when an institution has stopped asking whether quantum works and started asking how to put it to work.

One more thing, and to a European it is not a small thing. The systems DLR QCI will use were built by us and run in a data center we operate, in Germany. The workloads stay in the country. For a federally funded national initiative that is not a preference, it is a requirement, and it shows up formally in serious tenders now. It is also one of the reasons owning the full stack matters more every year. The machine, the software, the building it lives in.

One more thing, and to Europeans it is not a small thing. The systems DLR QCI will use were built by us and run in a data center we operate, in Germany. The workloads stay in the country. For a federally funded national initiative, that is not a preference, it is a requirement. It is also one of the reasons owning the full stack matters more every year: the machine, the software, the building it lives in. As Europe prepares its Quantum Act, this is the kind of capability it should be building: quantum systems made, run and used in Europe.

None of this happened alone. Bechtle has been our partner since 2024, and their Competence Centre for Quantum Technologies handles the onboarding and administration that make a platform usable by a large institution rather than merely accessible to it. Usable, not just accessible. That distinction is most of what Production Quantum means.

We are still early. I say that often and I mean it. But this is how the era actually arrives. Not with a breakthrough announcement. With a serious institution putting quantum into its everyday research infrastructure and getting on with the work, the way engineers always have.

Zuse liked to joke that he was too lazy to calculate, so he built a machine to do it. Underneath the joke is the entire idea of computing infrastructure: the machine exists so the people can think about the problem instead of the machinery. That is what DLR QCI just procured. Not qubits to admire, but infrastructure to think with. Eighty years apart, the same good idea.