Same way you do for planes, cars, etc. You have long(ish) life-cycles and some pieces that can be independent (e.g., same turbine engine on several plane models).
Boeing didn't want the time, expense, and hassle of certifying (= standardising) a new narrow-body aeroplane, so they continued to reuse the FAA type certificate (= standardised design) of the original 737 from 1966.
This meant they had to keep, inter alia, the short landing gear, which in turn made the wings lower to the ground, which forced them to position the new big engines ahead of the CG, which forced them to add the faulty MCAS computer, which killed all those people.
Admittedly the decision to use just a single sensor on said MCAS was due to systematic, decades long corruption and emasculation of the FAA.
Again, the standardization didn't cause the problem. Boeing's piss poor engineering culture did. There's no reason that they couldn't have built the plane how they wanted but in a way that didn't crash. Similarly, it's entirely possible that each of these nuclear reactors will be built with flexible designs per project that result in half of them melting down.
Safety and quality control is critical no matter what strategy they use