EDIT: since people post so many pics of themselves on facebook under all sorts of lighting conditions and angles, that might actually make face recognition somewhat feasible from a cell phone cam. of course, efficiently searching thru a corpus of millions of faces (each taken at several different angles) is an enormous technical challenge. i envision some app like Shazam being developed to recognize faces rather than songs ...
He told me that they built it from scratch and did not use OpenCV. Some parts were even written in assembly.
Im pretty sure that this is the only web application that I use that has had major parts of it written in assembly.
But for assembly optimization it is really not as useful as it used to be. Scalability is more about how to get sub-linear time complexity and efficient communication pattern. Nowadays c compiler can get fairly good assembly code and for low level optimization, human cannot compete with machine (how many people knows the particular cache line alignment trick on old core i7?). Multimedia instructions (SSE/MMX/3D Now) are useful but most of them can be done by function call instead of hand-crafted assembly.
There was research cited here a while ago on the high accuracy obtained by constraining the search space to within one's social graph (hundreds vs millions). While it might not identify the stranger at the bar, it might identify their companions and so on. Six degrees of separation in real-time.