The difference is that the engineer has a precise knowledge of all components, their interactions, and their behaviour under various stresses. The material scientist deals with more vague and uncertain properties, such as superconductivity or even the stability of the crystal structure itself, so the process relies much more heavily on the trial and error.