Irfan Siddiqi is one of the leading researchers in the field of quantum feedback and has done some really awesome experiments. I think developing new architectu

Irfan Siddiqi is one of the leading researchers in the field of quantum feedback and has done some really awesome experiments. I think developing new architectures for superconducting quantum chips is very important and there are many things that are still unresolved. John Martinis' group had (and maybe still has) a very open-source approach to this as well, they published most of the source code and custom-built electronics they use in their labs for controlling qubits (the architectures themselves are documented more in papers and theses). It seems their wiki is no longer reachable from the public web, shame. The quantum computing startups try to convey the impression that they solved all problems for building real-world quantum computers already, I think that many things are still open though (and so far none of the startups has really produced convincing evidence that they really have a working quantum computer, or even an interesting qubit chip). What I'd like to see from these startups are at least some basic measurements that characterize the quality of single-qubit manipulations, two-qubit gates and qubit readout. With that one could at least estimate what one could do with their quantum devices.