Bayesian posterior approximation is more accessible to practitioners than ever thanks to modern black-box software. While this software offers widely accurate approximation with minimal user effort, it is well known that certain posterior geometries remain challenging for standard approximation schemes. As such, research into alternative approximations continues to thrive. In these papers, it is common for authors to demonstrate that their new approximation works well by testing it on posterior shapes considered to be challenging or "wild.'' But the shapes are not always directly connected to a practical application where they might arise. In the present note, we provide examples of applications in the wild that give rise to some common benchmark posterior shapes.