Somehow I just now discovered this algorithm and thought it was cool so I took a shot at implementing it.
Nothing terribly fancy, probably only supports distinct numbers, also only using last element as the pivot so not doing anything super intelligent here.
If you're like me living under a rock, the quickselect algorithm can find the k-th largest (or smallest) item in an unsorted array. This can be used to compute medians in linear time (on average). Its implementation is extremely similar to quicksort where a partition is involved using a pivot and recursively traversed.