The state of stress during the bottom-up assembly of nanoparticles strongly correlates with the microstructure of dense nanoparticle aggregates therein. A range of interaction length scales exists in these dry granular systems spanning from particle-scale elastic repulsion to aggregate van der Waals cohesion; the competition among these interactions dominates athermal microstructural evolution under applied stress. In this work, structural optimization is employed to simulate the nano-mechanical physics of athermal densification and jamming. The translational and rotational motions of nanoparticles are optimized to static equilibrium. An initially sparse and random configuration of particles is compacted into a mechanically stable (i.e., jammed) state by densifying the system under various external-loading paths (e.g., hydrostatic, uniaxial, and shear). The resultant jammed structures and their responses to shear exhibit strong correlation with the strength of interactions in addition to particle shape [see Smith et al., Phys. Rev. E, 82, 051304 (2011)]. The structural information, such as particleparticle contact types and pore geometry of the heterogeneous media in these densified systems will aid in understanding energy transport for functional applications such as thermoelectric elements and battery electrodes.