Mortality due to organ failure presents a significant medical challenge with liver failure alone responsible for over 35,000 deaths per year in the United States. Currently, the only therapy that offers any significantly improved prognosis is organ replacement, but because the demand for transplantable organs far exceeds supply, only the most severe cases are referred for transplant. While engineered tissues offer the promise to alleviate the suffering imposed by organ failure, they continue to lack key features that would make them a viable treatment option. In particular tissues require a developed vascular network to deliver nutrients and oxygen and to remove metabolic waste products. This is particularly important in highly metabolic and complex tissues like the liver. The problem of vascularizing tissue constructs has been addressed using a variety of approaches, but to date those approaches have seen limited success due primarily to lack of scalability. Self-organization, where basic building blocks are built autonomously into larger structures, has been successfully employed in a variety of engineering applications. We propose that the same physical principles of self-organization that govern the development of river tributaries can be utilized to construct an efficient microvascular network in an adhesive particle suspension. Using cell-laden hydrogel modules as the basic tissue building blocks, we propose to construct a perfusion system in which the macroscopic system parameters can be tuned to create a vascular network that efficiently delivers nutrients to cells throughout the tissue to maintain long-term tissue function and viability. The overall goal of this project is to develop an approach to vascularizing liver tissue in vitro, where the tissue can be designed a priori and built from the ground up using basic tissue building blocks. We hypothesize that physical models of erosion can be applied to suspensions of cell-laden hydrogels and that these models will allow us to optimize the organization of a vascular network that is capable of maintaining cell function and viability. In order to achieve this goal, we propose: (1) to study self-organization in an adhesive suspension of microscale hydrogels under flow and (2) to optimize self-organization of microscale hydrogels for developing a phenotypically functional in vitro liver model. The result will be a fully scalable technique for designing complex tissue architecture from the ground up.