Stein's method is a general framework based on solutions of certain differential or difference equations for bounding the distance between the distribution of a random variable $X$ and some specified target distribution. In particular, it can be used to supply rates for various limit theorems and it is often able to handle dependence relations which more classical techniques cannot. In this talk, I will outline some of the main ideas in connection to normal and Poisson approximation, and will discuss other extensions if time permits.