Designable systems with huge numbers of controllable degrees of freedom are becoming a mainstay of machine design, from elastomeric shape shifting sheets and microscopic metamaterial robots to de novo protein design. State of the art inverse design techniques allow the organization of a system's degrees of freedom to achieve a shape transformation into a single target shape. However, to perform a function these systems must be designed to cycle between multiple configurations. The outstanding challenge is the organization of these degrees of freedom such that a system undergoes a sequence of configuration transformations to perform a function, transforming a material into a machine. I will describe a series of strategies that advance these goals. First I will show how elastic systems with noninteracting local degrees of freedom can be designed to transform and cycle between multiple configurations in response to a sequence of global actuations. I will then describe a novel universal kirigami pattern that by explicit local control of all of its degrees of freedom can smoothly transform between arbitrary shapes. Finally, I will ask how do we robustly organize multiple degrees of freedom with long ranged interactions to respond to simple controls, such that a system snaps between multiple states. I will describe a framework revolving about bifurcations of multiple equilibria for the design of such systems, and demonstrate its implementation in a magneto elastic system. These novel design approaches are especially relevant in microscopic systems. Indeed, microscopic machines should not be miniatures of their macroscopic counterparts, because the relevant physics in these scales is different. These three new design paradigms rise to answer Feynman's challenge in his "plenty of room at the bottom" lecture calling for novel strategies in the design of microscopic systems.