Small RNAs are an exciting class of therapeutics with significant untapped therapeutic potential, due to their ability to affect cell behavior at the RNA level. However, delivery of RNA is a challenge due to its size and labile nature. Extracellular vesicles (EVs) are promising as delivery vehicles due to their natural role as physiological intercellular microRNA transporters, and research has shown that EVs have significant advantages compared to competing technologies such as lipid nanoparticles. Specifically, EVs more readily transport through biological barriers, deliver RNA more efficiently, and are less immunogenic. However, intrinsic microRNA content in EVs is low and thus active small RNA loading strategies are needed to enable therapeutic use. Consequently, a variety of small RNA loading methods for EVs have been developed. These include endogenous and exogenous approaches. Exogenous approaches, in which EVs are loaded directly, have been shown to enable loading of hundreds to thousands of small RNAs per EV, but they are not readily amenable to scalable production processes. Endogenous approaches, in which EVs are loaded by upstream manipulation of the producer cell, are compatible with large scale EV production, but loading by these approaches is inconsistent and has scarcely been quantitatively analyzed. The work in this dissertation is focused on enabling small RNA therapeutics via EV delivery. The lack of an ideal small RNA loading approach for EVs is addressed by tackling important issues of both endogenous and exogenous loading. First, the loading capacity of several common endogenous loading methods was optimized and quantitatively analyzed. Additionally, new approaches to endogenous small RNA loading involving genetic manipulation of the RNA structure and the microRNA cellular processing pathway were developed and evaluated. Finally, exogenous loading via sonication was applied to enable delivery of a novel microRNA combination that was identified via a rational selection process. This combination of miR-146a, miR-155, and miR-223 was found to have potentially synergistic anti-inflammatory activity, and EV-mediated delivery of the combination opens the possibility for therapeutic application in inflammatory diseases and conditions such as sepsis. Overall, this work both improves understanding of current techniques for small RNA loading into EVs and opens new opportunities for advanced strategies, bringing EV-based small RNA therapeutics closer to clinical application.