This is nothing more than an engineering challenge, looking at it seriously. Cooling? an IR barrier in the reactor to stop IR from reaching the cells should so

This is nothing more than an engineering challenge, looking at it seriously. Cooling? an IR barrier in the reactor to stop IR from reaching the cells should solve about 80% of the issue - assuming you can make something transparent at the right wavelengths for the cells - and capable of withstanding the heat of the reactor. The other 20% is already doable with current technology, and we do it from large-scale solar farms to single-family homes - water tubing on the backs of the devices to a radiator or storage tank. Hell, while using them to harvest light energy, that water loop can be used to harvest heat energy, thus improving the overall system efficiency!