• Our data harvesters captures public profiles from several 1,000 data sources including Google profiles, Google publications, Indeed profiles, Xing, GitHub, Stack overflow, Patent databases etc.. This data gives us the names, location, companies, roles and skills of individuals across the world. We use this data to capture sample(Representative) for every job family for a given location
  • For the sample profiles we run our name matching algorithms. Name matching algorithm gives us an allocation of ratios of profiles for the skill/job categories
  • We also seek public reports from census and other 3rd party sources to validate the data.
  • Based on the public comparison data we then use floating and fixed ratio algorithms to estimate the gender ratios of job/skill categories which are validated