Large scale structures such as cosmic voids and galaxy clusters are a crucial window into cosmology and fundamental physics. Accurately estimating properties of cosmic structures such as their cluster mass and void size has historically been difficult, however, since the underlying dark matter distribution cannot be directly observed. Many of these challenges can be overcome, however, by using Field Level Inference with the Bayesian Origin Reconstruction from Galaxies (BORG) algorithm. This approach uses forward modelling to sample the posterior distribution of possible initial conditions, which can then be resimulated in order to study the properties of the dark matter distribution in the local Universe. In this talk, I discuss recent progress in using posterior resimulations to estimate the masses of galaxy clusters, and in using posterior simulations with inverted initial conditions to construct an antihalo void catalogue.