Stochastic perturbations have become increasingly important in fluid dynamics, both as a modeling tool for uncertainty and turbulence and as a mechanism influencing analytical well-posedness. While stochastic fluid equations are by now well studied from a theoretical perspective, their structure-preserving numerical approximation remains a major challenge. In this talk, I discuss how stochastic effects can be incorporated into numerical methods for inviscid fluid models, focusing on the stochastic Euler equations. I outline the key difficulties arising from randomness, such as stability, entropy control, and consistency with probabilistic solution concepts, and explain how entropy-dissipative discontinuous Galerkin and finite volume methods can be adapted to address these issues. The approach is guided by the notion of dissipative martingale solutions, which provides a natural analytical framework for convergence in the stochastic setting. Numerical experiments illustrate how stochastic forcing influences solution behavior and demonstrate the robustness of the proposed discretization. Overall, the talk aims to highlight general principles and strategies for designing reliable numerical methods for stochastic fluid equations.