When it comes to nuclear, that's a problem the world over. Out of practice of designing as well. Look no further than the mess that is the EPR. On its face it seems like a boondoggle, but it makes a lot more sense when you look back at the history of nuclear construction: from 1971 to 1993, France was continuously building plants, first the 34 900MW CP-series (1971 to 1988) then the 20 1300MW P-series (1978 to 1993). The CP-series took 5-6 years to build and the P-series a bit more (6-7). From the mid-70s to the mid 80s, there were a dozen plans being built at any time, and they were being completed.
Then came the N4, that thing had serious teething issues, the first model started construction in 1985 but took 12 years to complete. By the time the 4th was completed, it was down to 7 years. But the 4th was the last N4 to be built. It's unclear to me whether the N4 slowdown was intended all along or whether the teething issues made them take a look at construction and pump the brakes, but in the space of 5 years the country went from building 12 plants at a time to building 5, then 4, then 3, then 2, then 1.
N4 at least overlapped some with the tail end of the CP-series and the middle of the P-series, but the first EPR construction was started in 2007, 5 years after the last N4 (Civaux 2) was completed. I can't even imagine the loss in institutional knowledge over that span.