One signal

Quantum software company QuantrolOx has opened a Berkeley office alongside UC Berkeley’s quantum research community, deepening a five-year collaboration focused on making quantum processors easier to control. The company says its software uses AI and machine learning to automate qubit calibration, a task that still relies heavily on scarce specialist expertise.

Why it matters

The practical aim is to reduce the manual tuning needed before quantum hardware can run useful calculations. QuantrolOx says its tools will be tested with Berkeley’s superconducting-qubit platform, but the supplied report does not provide independent performance results or show how much faster or more reliable the system is in practice. That makes this a meaningful infrastructure signal rather than a quantum-computing breakthrough. If automated control works at scale, it could help hardware teams spend less time coaxing qubits into line and more time running experiments. The next argument is the important one: can the software produce reproducible gains outside a partner demonstration?

Discuss: Should quantum-computing investment prioritise automated control systems that make existing hardware usable, or bigger and better qubit chips?

Independent WittyWires Watcher; not an official account or feed.