Uber's technology is reportedly 'hanging by a thread' - but the company's CTO is getting it together

The technology that powers Uber's app is "hanging by a thread," according to a new report from The Information's Amir Efrati.

Uber has grown so rapidly over the past few years that its technical infrastructure has struggled to keep pace, according to The Information. Efrati notes, "Uber is growing so fast that every three months, what was peak traffic becomes its average traffic."

In addition, Uber has two different systems - a Node.js system that dispatches cars and a Python system that's a backend system for fare calculations and customer emails. The engineers in charge of these systems have been "at odds," which has created friction, according to The Information.

But the company is working on its engineering woes with help from Thuan Pham, who previously worked in ad tech and at VMWare before coming to Uber toward the end of 2013 to serve as the company's CTO.

Uber CEO Travis Kalanick hired Pham only after calling him on the phone every day for two weeks to quiz Pham on recruiting and management, Pham told Fast Company earlier this month.

Under Pham's watch, Uber is working out its technological issues. More than anyone else, he knows there's problems - in an email to his staff obtained by The Information, Pham said one outage "reflected an amateurism with our overall engineering organization, its culture, its processes, and its operation."

The company's engineering staff has grown to 1,200 - a quarter of Uber's workforce - from just 400 people. Pham's goal, he tells The Information, is to reduce Uber's downtime on the app to just 50 minutes a year - to hit 99.99% reliability, so the app is nearly always working.

Uber doesn't use a public cloud provider for hosting its app, instead relying on third party providers to handle its servers. Under Pham, Uber has hired people to take care of network engineering and handling the servers.

The engineering team at Uber appears to be learning from their mistakes as they grow. The company took a big bet by setting up servers in a new data center on Halloween - one of Uber's biggest days of the year - last year. This ultimately caused outages and Uber was forced to go back to its old hosting provider. But, Pham tells The Information, "we failed early; we learned fast and as a result; New Year's Eve was flawless, with around 1.7 million trips."

We reached out to Uber for comment on this story and will udpate if we hear back.

You can read the full story from The Information - which illustrates the challenges of engineering for fast-growing, $50 billion Uber - here.