Current methodologies for stress inversion from microseismic focal mechanisms require the assignment of events to a regular grid and then solving for the stress state at each grid node. This approach can lead to irregularities in the solution because some nodes may contain few or even no events. To address this issue, we modified the algorithm to solve for stresses on an irregular (unstructured) grid. We first use the -means algorithm to split the data into suitably sized groups. The centroids of these groups are then considered as the nodes of an unstructured grid, and we simultaneously solve for the stress state in each group using damped inversion. To account for the irregularity of the unstructured grid, we use the reciprocal square distance between nodes as weights, as opposed to the existing method in which a weight of unity is assigned between adjacent nodes on a regular grid. Focal planes are selected from the auxiliary plane using the fault instability criterion. The method is first applied to synthetic data sets in which we simulate and subsequently invert for the stress field around a mode-I fracture at depth, in a strike-slip and in a normal faulting stress regime. Results indicate a stress orientation error of 10° and a stress ratio error between 1% and 10%. We then consider focal mechanism data from an unconventional shale play in the Vaca Muerta Formation in Argentina, and our results suggest the presence of a preexisting strike-slip faulting stress regime. We also find that the unambiguous focal plane picks suggest that the apparent dip-slip focal mechanisms are indeed dip-slip movement along subvertical natural fractures, which correlate well with image log data. We suggest that these dip-slip events are caused by stress changes induced by the opening of the hydraulic fractures.