Recent models of hot cores have incorporated previously-uninvestigated chemical pathways that lead to the formation of complex organic molecules (COMs; i.e. species containing six or more atoms). In addition to the gas-phase ion-molecule reactions long thought to dominate the organic chemistry in these regions, these models now include photodissociation-driven grain surface reaction pathways that can also lead to COMs. Here, simple grain surface ice species photodissociate to form small radicals such as OH, CH3, CH2OH, CH3O, HCO, and NH2. These species become mobile at temperatures above 30 K during the warm-up phase of star formation. Radical-radical addition reactions on grain surfaces can then form an array of COMs that are ejected into the gas phase at higher temperatures. Photodissociation experiments on pure and mixed ices also show that these complex molecules can indeed form from simple species. The molecules predicted to form from this type of chemistry reasonably match the organic inventory observed in high mass hot cores such as Sgr B2(N) and Orion-KL. However, the relative abundances of the observed molecules differ from the predicted values, and also differ between sources. Given this disparity, it remains unclear whether grain surface chemistry governed by photodissociation is the dominant mechanism for the formation of COMs, or whether other unexplored gas-phase reaction pathways could also contribute significantly to their formation. The influence that the physical conditions of the source have on the chemical inventory also remains unclear. Here we overview the chemical pathways for COM formation in hot cores. We also present new modeling results that begin to narrow down the possible routes for production of COMs based on the observed relative abundances of methyl formate (HCOOCH3) and its C2H4O2 structural isomers.