The year before Henke joined, the opening of LinkedIn’s second data center had gone so badly that executives almost had to call the board in the middle of the night to say the project had failed because of a database bug, says Igor Perisic, LinkedIn’s former chief data officer. Office TVs displaying usage statistics would fill with squiggly red lines, denoting failed sign-ups due to overloaded servers, Nishar says. “The success definitely has underpinnings in the technologies that we invested in at that time,” he says. Without the focus on “site up,” LinkedIn may have struggled to become the giant it is today, says Deep Nishar, its product head from 2009 to 2014. Company veterans recall the project as one of the most pivotal in the company’s 20-year history. That could choke a healthy business that had sought profits early on to win over venture capitalists who were still spooked by dotcom busts.ĭuring the next few weeks of around-the-clock coding, the engineering team rebuilt LinkedIn’s technology from the ground up, significantly reducing downtime. That was standard practice in startups, but the network was now big and important enough that unreliability threatened to stunt its growth and scare off clients, Henke thought. The weekly downtime, which included an office happy hour to lubricate a long night’s work, stemmed from LinkedIn’s scrappily built technology. Clicking on a profile would summon the “ wizard of ,” a jaunty, staff-wielding mascot akin to Twitter’s famous “ fail whale.”

Recruiters paying to trawl the website for candidates were forced to twiddle their thumbs. The fast-growing professional social network had about 50 million users, but every Thursday afternoon it went completely offline as engineers launched new features and fixed bugs. “What the heck is wrong with all of you?” was the sanitized gist of it. In 2009, LinkedIn’s new engineering chief, David Henke, assembled his full crew of coders and managers for the first time and fired off tough questions.

Linkedin jobs near me