18 November 2015
Eddie Vivas is in an odd position. His job is to apply mathematical formulas to LinkedIn’s millions of personal profiles to match the right potential job candidate with the right job opening. Yet he also admits that no algorithm could have predicted that a high-school dropout from Florida such as himself would have become “head of talent solutions” at one of the behemoths of Silicon Valley.
That uneasy truth puts the boyish-looking 30-something in the middle of an ideological battle over the future of recruitment.
On the one hand, LinkedIn is deploying Mr. Vivas’s computer science to help employers find new staff who will be a good fit – either because they are already connected to people there, or because their experience, skills and education make them good replicas of favorite employees. “We have the ability for you to just tell us ‘here are a group of people who I think are amazing’ and we’ll find others just like them on the platform,” he says.
But a growing number of companies are moving in the opposite direction: they are actively hiding information on candidates’ CVs from their recruiters, so that they do not end up with homogenous workforces where everyone looks the same, sounds the same, and has the same college scarf.
EY, a business services company, for example, now selects entry-level candidates for interview based only on their performance in online tests, and does not give interviewers any background information (such as work experience or academic grades).
Politicians – increasingly fretful about social mobility – are warming to the idea too. Last month David Cameron, UK prime minister, announced that a group of public and private sector employers would implement “name-blind” recruitment for young people to try to combat racial discrimination in the hiring process.
Does Mr. Vivas see the tension between these trends? “Yes and no,” he says. The tools he has designed for LinkedIn do not necessarily perpetuate homogeneity, he argues. “My background doesn’t represent the LinkedIn background at all, and me being able to bring more people that I know into our organization, I think is a really good thing”. Similarly, he could use the tools to find more people who are just like his “amazing” engineering partner Annabel Liu.
<The tabular content relating to this article is not available to view. Apologies in advance for the inconvenience caused.>
“The issue of diversity isn’t at the tool level, it’s at the organizational level,” he says, adding, “I actually don’t think any of these tools, in themselves, are the cause of that.”
LinkedIn is also thinking about how it could use its vast amounts of data to promote diversity, by showing how diverse companies are relative to their peers and to the labor market as a whole.
“Without data, there’s a lot of plausible deniability,” he says. “But diversity of ethnicity, color and race is only part of what I think is important to solve. I think diversity of thought is really important too. I bring a very different perspective to the table than what a Harvard grad will.”
Yet Mr. Vivas’s path to LinkedIn from his boyhood home in Jacksonville, Florida (“as far away from Silicon Valley as you can get,” he says with an impish smile) also demonstrates the limits of what data can do.
After he dropped out of high school at 17, he moved to Chicago and worked some “tough jobs” until he met his first business partner – a man who took a chance on the youngster, even though any algorithm would have told him to steer well clear. That started Mr. Vivas on the road to where he is now, having sold his last company, Bright.com, to LinkedIn last year for $120m.
“Relationships matter: someone was willing to take a risk because they believed in me, even though my qualifications didn’t back anything up at all,” he says.
“Data’s incredibly powerful, but data alone can’t help you hire great talent. Robots and algorithms are not going to replace recruiters. And actually, I used to believe that, by the way – I used to believe that algorithms could identify exactly who you should hire.”
Real life has corrected that view.
“Once somebody has spent time working in the recruitment space, and understands all the intricacies that actually exist in dealing with human beings . . . they learn that data’s a piece of the puzzle, but it’s not everything.”
Mr. Vivas subverts Silicon Valley stereotypes in other ways too. He may be a 30-year-old technology entrepreneur, but he does not know how to code.
“I know what’s possible with it, but coding – actually programming – I don’t know how to do.”
In fact Mr. Vivas thinks it is troubling that policymakers are trying to make all schoolchildren learn how to code.
“I think we’re moving to a world where people’s understanding at a high level of how these systems are built is going to be important, but the actual act of coding is not what we need everybody to start focusing on doing,” he says.
“Coding is kind of like an art – just because you can get somebody to spend the time doing it, if they don’t know what they’re doing, you’re going to get a mess.
“The world doesn’t need more bad coders.”
Copyright The Financial Times Limited 2015
By Sarah O’Connor, Source: FT.com
(c) 2015 The Financial Times Ltd. All rights reserved. Please do not cut and paste FT articles and redistribute by email or post to the web.