The current generation of fast-running whole-core, coarse-mesh nodal methods will remain the most practical methods for reactor depletion in the near term, despite considerable interest and effort in high-fidelity methods taking advantage of increased computational resources. In this work, the TRITON/PARCS/PATHS two-step sequence was used to model the BEAVRS benchmark. Additionally, cross-section uncertainties were propagated throughout the procedure using Sampler/TRITON/PARCS/PATHS. Comparisons were made between the predicted core response and the reported values for low power physics tests, critical boron concentrations, and detector responses. It is observed that the TRITON/PARCS/PATHS two-step procedure shows good agreement to the measured quantities, with an average critical boron discrepancy of 23 ppm and 18 ppm for the first and second cycles, respectively. The cross-section uncertainty is observed to have a large effect on the criticality, with an average standard deviation of 40 ppm and 45 ppm for the critical boron concentrations in the first and second cycles, respectively. In addition, comparisons to the measured flux maps show consistent agreement, with an average radial Root Mean Square relative error (RMS) of 3.25% for the first cycle and 3.06% for the second cycle. The effect of cross-sectional uncertainties is observed to be much weaker on the flux comparisons for each cycle as the average standard deviation of the radial RMS values were 0.38% and 0.35%, respectively.