The title is technically correct; however, it seems like link bait once you get to page two of the article and see: "Bridenbaugh told ABC News that he believes the design flaws that prompted his resignation from GE were eventually addressed at the Fukushima Daiichi plant."
> he believes the design flaws that prompted his resignation from GE were eventually addressed Anyway, a design that blows up when power is cut because it cannot cool itself down to safe levels is... weird.
In short, even after reactor shutdown, substantial heat is still produced, so some cooling is needed. Substantial as in, up to some 7% of normal output, early after the shutdown. That's mostly due to residue fissible elements -- byproduct of normal operation.
The explosion was not inside reactor pressure vessel, but in the housing. Top parts of fuel rods got exposed out of coolant water at some point and heated up considerably. When water level was raised back again, heat caused release hydrogen. Upon subsequent steam venting (necessary in case of overheating), the hydrogen combusted in the air in reactor housing (i.e., building, not the pressure vessel itself). The pressure vessel itself contains no air to avoid explosion.
Compare a very early mishap with a nuclear pile (not reactor): http://en.wikipedia.org/wiki/Windscale_fire -- where the Brits tried to extinguish smoldering graphite & very hot fuel with water, and there was risk of hydrogen explosion (fortunately avoided).