This paper proposes a variance estimator for the OLS estimator and nonlinear estimators such as logit, probit, and GMM, enabling cluster-robust inference in the presence of two-way or multi-way non-nested clustering. The estimator extends the standard cluster-robust variance estimator for one-way clustering and relies on similar weak distributional assumptions. It is easily implemented in statistical packages like Stata and SAS, which already offer cluster-robust standard errors for one-way clustering. The method is demonstrated through Monte Carlo analyses for a two-way random effects model and a placebo law example, as well as applied to empirical studies with two-way clustering. The paper also discusses practical considerations, such as small-sample corrections and handling of negative elements in the variance matrix. The approach generalizes to multi-way clustering and can be adapted for weighted least squares and feasible GLS estimators.This paper proposes a variance estimator for the OLS estimator and nonlinear estimators such as logit, probit, and GMM, enabling cluster-robust inference in the presence of two-way or multi-way non-nested clustering. The estimator extends the standard cluster-robust variance estimator for one-way clustering and relies on similar weak distributional assumptions. It is easily implemented in statistical packages like Stata and SAS, which already offer cluster-robust standard errors for one-way clustering. The method is demonstrated through Monte Carlo analyses for a two-way random effects model and a placebo law example, as well as applied to empirical studies with two-way clustering. The paper also discusses practical considerations, such as small-sample corrections and handling of negative elements in the variance matrix. The approach generalizes to multi-way clustering and can be adapted for weighted least squares and feasible GLS estimators.