Most many body methods for solving the Schrodinger Equation - perturbation theory, coupled cluster theory, Green’s function theories, etc. - are deterministic in nature. While deterministic methods can be highly accurate, they often scale steeply with system size. In this talk, I will discuss a suite of new quantum Monte Carlo methods my group has recently developed that use stochasticity - randomness(!) - to solve a variety of ground state and finite temperature problems in quantum mechanics difficult to approach using deterministic techniques. In particular, I will highlight stochastic variational methods we have developed that enable us to explore the ground state physics of single and multiorbital Hubbard models via simulations that near the thermodynamic limit. I will also introduce a new ab initio finite temperature Auxiliary Field Quantum Monte Carlo algorithm that opens the door to highly accurate finite temperature simulations of solids. These algorithms will ultimately enable the study of materials with the rigor of chemical accuracy at a cost nearing that of Density Functional Theory.