Biomolecular machines are central actors in essentially any major cell biological process. Their successful function requires effective energy conversion and handoff between diverse mechanical components, and symmetry-breaking to achieve directed transport. It seems plausible that evolution has sculpted these machines to efficiently transmit energy in their natural contexts, where energetic fluctuations are large, nonequilibrium driving forces are strong, and biological imperatives require rapid turnover. But what are the physical limits on such nonequilibrium efficiency, and what machine designs actually achieve these limits? In this talk, I address how to drive such noisy systems rapidly and efficiently from one state to another, and how to allocate nonequilibrium driving forces among the steps of a machine cycle to maximize its throughput. These theoretical results find confirmation in experiments, have immediate consequences for the design of single-molecule experiments and molecular simulations, and provide nontrivial yet intuitive implications for the design principles of molecular-scale energy transmission.