Digital Fast Company

Data Scientists Compete For Glory (And To Get Noticed By HR)


Jennifer Alsever

June 19, 2014

Big corporations are convincing some of the hottest young coding talents to compete for their attention. By hosting crowdsourced competitions, corporations can pit hundreds of technical people against each other to crunch information and solve business problems. The winners land job interviews.

Facebook hosted three competitions on, which has a network of 180,000 academics, working data scientists, and hobbyist statisticians. One of those competitions challenged candidates to show “creativity, open-mindedness, and tenacity,” by recommending missing links in a social network. Yelp, meanwhile, asked competitors how many “useful” votes an online review might receive, and Walmart recently asked people to project sales at 45 stores and their various departments in light of holiday markdown events.

The idea: tap into a network of thousands of smart people, test skills objectively, and showcase the kind of interesting problems people can solve if hired. That last part is key, since demand is growing for data scientists, or people skilled in statistics, as well as computer programing and business. McKinsey Global Institute recently estimated a 140,000 shortfall of people with those skills.

“Competition is fierce,” says data scientist Claudia Perlich, who became an all-star in the industry after winning two similar competitions. But why are some of the most sought-after talents fighting each other just for the chance at an interview? “It’s all for the glory,” Perlich says. Participants want to win for the sake of winning, she says, not because they see it as a way to get noticed by HR.

Of course, Perlich says participants do share stories of companies adding an extra $100,000 onto salaries to woo top candidates from competitors. In fact, Kaggle made the leaderboards of its hiring competitions private after a competitor monitored Facebook’s competitions and attempted to poach the winner by offering them a job interview before Facebook did, says Kaggle CEO and founder Anthony Goldbloom.

Companies see huge competitive advantages in being able to mine and make sense of the vast amounts of digital information being produced online and inside corporate walls. But for many hiring managers, assessing a job candidate’s cultural fit can be much easier than determining their technical skills: A “Best in Show” medal makes it easy to identify the elites.

The crowdsourced competitions, Goldbloom says, offer an objective way to measure technical skills and offer a taste of the kinds of work a person might do at a company. “It becomes a job advertisement,” he says. Kaggle originally started, however, as a way for companies to tap crowds to solve data-mining problems and offer cash prizes to winners.

Last month, Walmart’s hiring competition drew 694 competitors. There is good reason: With sales approaching half a trillion dollars and more than 250 million customers in stores every week, the company is collecting huge amounts of data. “To a data scientist,” Goldbloom says, “Walmart is like Disneyland.”

Great ! Thanks for your subscription !

You will soon receive the first Content Loop Newsletter