Polymers that exhibit no appreciable long-range order are important for a variety of real-world applications ranging from water filtration membranes to ballistics resistance materials to structural materials. Molecular packing structures in these materials can control important properties but are difficult to measure due to their fine size scales and irregular packing patterns. These structures can be predicted using simulations, but high-quality experimental data is required for determining the most highly probable candidate structures. To this end, we utilize results from solid-state NMR measurements for biasing simulations towards the most highly probable structures. Several solid-state NMR interactions can be used to probe local structure: nuclear dipole-dipole strengths for internuclear distances, electric field gradients for atomic bonding arrangements involving quadrupolar nuclei, and magnetic shielding tensors for atomic bonding geometries. In this work, we utilize (1) atomistic molecular dynamics simulations of large boxes of atoms (>1000) biased by internuclear distance measurements (i.e. REDOR), and (2) ab initio calculations of small clusters of atoms (<100) biased by chemical shift tensor and nuclear quadrupolar tensor measurements (i.e., “NMR Crystallography.”)