> The moment you hand them something fuzzy, though, like ...
  > “we should probably think about scaling”,
  > that’s when you see the difference.

Senior engineers should ask, "but do we need scaling? And if it does, how much needed now and future?" But I've seen a lot of seniors who jumped to implementing an unnecessarily complicated solution without questions, because they don't think about it too much, want to have fun, or just don't have energy to argue (I'm guilty myself).