The startup analyzed millions resumes and job descriptions, and used HR recruiters evaluations of the candidates and jobs to develop an algorithm that predicts compatibility and more.
Mashable, Sarah Kessler, Jun 19, 2012
A new job board called
Bright aims to kill resume overload with a little bit of data science.
Instead of encouraging candidates to apply for as many jobs as possible, Bright promises to cut down on unnecessary resume exchanges by showing both parties what is the best fit for them.
To do so, the startup analyzed 2.8 million resumes submitted in response to more than 2.1 million job descriptions on its beta site. Hundreds of HR recruiters looked at the resumes and rated them based on compatibility with their corresponding descriptions.
Meanwhile, a team of about 15 data scientists and engineers tracked the ratings and used them to develop an algorithm that not only predicts such compatibility, but also learns more about a specific employer's preferences as that employer uses the site. Bright uses it to assign each job seeker a score for each job.
Like competing job site
Path.to, it feels a little like a dating site.
Job seekers can see their scores for each job while they're browsing the site. And even if they search for the wrong keyword, they'll be given jobs that fit their skills. If a landscaper searches for the word "engineer," for instance, Bright will turn up witty calls for "plant engineers" but leave out the jobs that require an engineering degree.