Unlike the meteorologist, the developer has the ability to bend reality a bit when it comes to estimates, but they can never bend reality in a way which makes it worthwhile. Cutting corners is almost always a mistake. It's a very short term move which may only make sense if the business is collapsing and looking for a quick exit.
You could spend 15 years cutting every corner imaginable as a self-professed genius 1-man engineering team until everything becomes unsustainable, feature work is impossible due to frequent fires, the business is faltering, and you can't even understand your own code any longer. Then, quit as soon as a new hire comes in so you can't be held accountable for your quagmire by other engineers. I don't really understand why someone would do this, but I recently got to see it first hand. ;)