The biggest overarching point (that I found most interesting) was the shift in problem space from an "engineering problem" to a "statistics problem" since there

The biggest overarching point (that I found most interesting) was the shift in problem space from an "engineering problem" to a "statistics problem" since there were far too many unknowns to solve effectively with engineering principles alone.