Yeah but NASA are fantastic engineers. Their interface design and validation are orders of magnitude ahead of anybody else.
NASA didn't design the LEM, Northrop Grumman did. Spacecraft are designed by aerospace companies (like Northrop-Grumman, Boeing, Rockwell, and now SpaceX), and then NASA picks the design they like best. The best engineers are typically at the private companies because the pay is better than at government run NASA.
Consider the first shuttle flight. [...] And it worked first time. They were hot at the time, coming off the experience of Apollo.
Well, the first space shuttle, the Enterprise, never went to space. It's easy to have a successful first flight when you have the resources to build a full size scale model to 'test' with. And they weren't coming hot off Apollo; the space shuttle was about a decade later.
The most complex and unlikely machine (pretty much) ever built.
They made it needlessly complex. This is why they have had, and continue to have, so many problems. The designers promised several launches each month and a payload cost in $50-$100 per pound range.
The scientific community at the time said much the same things about the shuttle design that they currently say about the ISS; that it's too much money for too little return. Some even go so far as to suggest these overly-complex plans, pushed on the unsupportive science community are essentially aerospace company welfare.