Former Netflix Data Scientist Finds Smaller Data Sets “Less Daunting” for Fashion Algorithms
Eric Colson, Chief Analytics Officer of StitchFix, stopped by theCube to talk with hosts John Furrier and Dave Vellante about their recommendation engine. Often referred to as Pandora meets Zappos, StitchFix is powered by a recommendation engine for purchasing clothes (full interview below).
But wait, hold your horses. How’s that again?
“Algorithmically-chosen clothing. So it’s eCommerce,” says Colson. “We offer apparel items, but the customer doesn’t pick them out. We will choose on behalf of the customers based on their preferences and ship the merchandise right to their home, where they don’t have to keep anything they don’t want. If they decide they love it they can pay for it and send us the money. But they’re under no obligation.”
That’s right – the recommendation engine supported by human stylist pre-determine what will be sent to you based on your style, likes and even your Pinterest page. “It comes down to having very relevant merchandise for the customers, so that’s where the algorithm kick in…with just a touch of human interaction,” said Colson when he stopped by.
What once started as an algorithm-only idea, StitchFix quickly realized that the human element is needed. As explained by Colson, they offer two very different skills. StitchFix is a perfect blend of human and machine interaction. “Combination of humans and algorithms are not just complimentary but reinforcing,” said Colson. His team aims to create “elegant code.” Right now they do their discovery in R, and then rewrite some of the code to get it to work in a production system. The biggest challenge they face in their “elegant code” is finding a way to get it to work without the rewrite step.
Prior to joining Stitch Fix, Colson was the Vice President of Data Science & Engineering at Netflix, a company that is well known for its recommendation and targeting engine. Furrier ending the interview with a question that tied into Colson’s past, asked which data sources Colson is using to mine that kind of targeting?
“In terms of technologies, we don’t have the scale challenges that a Netflix has. The data is not unwieldy. So it means we get to put more of our attention toward the analytics the math, the modeling we do to build good predictive engines. …the volume is not daunting.”
StitchFix. Tackling your style for you, so that you don’t have to do anything but look good and take all of the credit. See Colson’s full interview below.
Since you’re here …
… We’d like to tell you about our mission and how you can help us fulfill it. SiliconANGLE Media Inc.’s business model is based on the intrinsic value of the content, not advertising. Unlike many online publications, we don’t have a paywall or run banner advertising, because we want to keep our journalism open, without influence or the need to chase traffic.The journalism, reporting and commentary on SiliconANGLE — along with live, unscripted video from our Silicon Valley studio and globe-trotting video teams at theCUBE — take a lot of hard work, time and money. Keeping the quality high requires the support of sponsors who are aligned with our vision of ad-free journalism content.
If you like the reporting, video interviews and other ad-free content here, please take a moment to check out a sample of the video content supported by our sponsors, tweet your support, and keep coming back to SiliconANGLE.
