||At the macroscale, fluids are typically modeled using the traditional Navier-Stokes equations, in which the underlying fluid particles are abstracted as continuous fields. While this approach is computationally efficient, it neglects the effects of thermal fluctuations. At the microscale, methods such as molecular dynamics (MD) explicitly simulate the underlying particles, capturing these effects at the cost of much greater computational expense. In between these two scales (i.e. the mesoscale), a range of problems exist which are small enough that thermal fluctuations play an important role, but too large to efficiently simulate using particle-based techniques. At this scale, fluctuating hydrodynamics (FHD) provides an approach which is relatively computationally efficient, but still incorporates the effects of thermal fluctuations. The FHD approach has been successfully applied to this type of problem. In this talk, I will describe how the FHD approach can be extended to two classes of stochastic chemical systems--reactive microfluids and heterogeneous catalysts--where both hydrodynamics and chemistry play essential roles. First, to construct numerical methods for reactive microfluids, I will explain how to incorporate various stochastic chemistry descriptions to the stochastic PDE description of FHD. Second, for small-scale heterogeneous catalyst applications, I will describe a hybrid simulation methodology that combines both the kinetic Monte Carlo (KMC) approach for surface chemistry and the FHD approach for hydrodynamic transport in the gas phase. I will also explain high-performance computing (HPC) implementations.