2 Comments

Summary:

Facebook is hosting a Kaggle competition in order to identify candidate for a data scientist position. Résumés are so passé when you can just have applicants prove their skills first.

Facebook is looking to hire a new data scientist, but rather than sift through résumés it’s planning to sift through algorithms. The company is turning to the data-science competition platform Kaggle to host a contest that will determine who gets brought in for an interview. This is the third time Facebook has used Kaggle to identify possible data scientists.

The challenge this time around involves text mining. Specifically:

“This competition tests your text skills on a large dataset from the Stack Exchange sites.  The task is to predict the tags (a.k.a. keywords, topics, summaries), given only the question text and its title. The dataset contains content from disparate stack exchange sites, containing a mix of both technical and non-technical questions.”

The competition began at 4:00 p.m. UTC on Friday and ends at 11:59 p.m. UTC on Dec. 20.

In a hiring environment where demand outstrips supply — especially quality supply — it’s arguable that anyone with aspirations of becoming a data scientist should be on Kaggle, Facebook challenge or not. Kaggle now boasts more than 100,000 members, and standing out among this group can be a good feather in one’s cap. At the very least, it’s a good chance to gain experience working with real-world datasets and possibly win some prize money in the process.

Feature image courtesy of Shutterstock user Simon Bratt.

  1. M. Edward Borasky Saturday, August 31, 2013

    1. The long time till the ‘contest’ closes tells me this isn’t really an ‘open position’. There’s no urgency to fill it. Facebook is trolliing.

    2. I seriously doubt if this test case (Stack Exchage text mining) is relevant to Facebook’s business. Facebook is an advertising business, and the kind of data science they need to boost ad revenues isn’t what this contest tests.

    So while 100,000 Kaggle ‘contestants’ are hacking away at this test, the likely hire is networking, trying to understand what Facebook’s *real* data science problems are, and probably publishing in the open literature. He or she is probably also employed by a competitor or in a startup Facebook can acq-hire.

    Share
  2. Srihari Yamanoor Wednesday, November 27, 2013

    You can fight. Or just wait for the bubble to get bigger. Facebook, and every other tom and harry will hire even your dog as a data scientist. From that point you will have about 6 months to a year for the crash.

    Share

Comments have been disabled for this post