"The sites that people like more, they put in one group. The sites that people like less, they put in another group. Then they look at tons of metrics....they use those in a machine learning process to essentially separate the wheat from the chaff."
This reminds me of the work done by the highest paid programmers in the world - financial algorithms. No surprise finance people tend to hire search people and digital signal processing experts since's they're essentially taking a massive amount of data (share prices, etc) and finding patterns in it.