While resin 3D printing allows designers to fabricate complex 3D objects, the technology has not found widespread adoption in manufacturing as a result of slow print speeds, poor reliability, and cumbersome support structures. The last of these in particular waste material, require human labor, are tedious to remove, and damage surface finish, but are fundamentally necessary due to adhesion forces and a lack of control of fluid flow during the printing process. Current design for additive manufacturing (DfAM) industry standards do not seek to offset such forces; instead, they empirically call for reducing printing speeds and/or imposing cumbersome supporting structures. Injection continuous liquid interface production (iCLIP) is a recent approach capable of effectively nullifying such forces by injecting resin into the deadzone. The method has been demonstrated to date for the case of a single channel running through an object formed of rigid material. However, the possibility of innervating the growing object with multiple channels – engineered into the CAD design uniquely for every print by this fabrication approach – remains unexplored. In this work we described our computational modeling and design approach to accompany iCLIP, optimally innervating the part with channels to infuse resin into the deadzone. We detail our modeling approach for both single and multiple injection sites, and for Newtonian and non-Newtonian resins. After describing our hardware implementation to evaluate our approach, we provide experimental validation of our simulation-driven injection scheme, including using both rigid and elastomeric resins. We demonstrate such a DfAM approach can significantly increase print speed and reduce the need for supports in a user’s 3D model. In doing so, our approach promises to enhance the scalability of resin 3D printing and to hasten its adoption in real-world manufacturing settings.