Nova Scotia tech company snags wins at Retail Systems Awards in London, England
HAVE YOU EVER COMPLIMENTED a friend’s sweater and wondered how to get one just like it? There’s an app for that. Well, there are many, thanks to Slyce’s technology and their research and development efforts. Slyce was conceptualized in 2014 and helps brands market products in real time by creating a technology using artificial intelligence and deep learning and then licensing it to retailers.
“The most common place you would see our technology in action is if you download the Macy’s app or the Home Depot app and in the camera mode of those apps—that is actually Slyce under the hood,” explains CEO Ted Mann in a telephone interview with Atlantic Business Magazine. Slyce got their start in fashion using artificial intelligence (AI) to identify products and match them to retailer’s catalogues with brands like Neiman Marcus and J.C. Penney. As they built the technology over time, they expanded to home repairs with Home Depot and electronics with Best Buy.
In June 2018, Mann accepted the Mobile Solution of the Year award on behalf of the Slyce team at the 13th annual Retail Systems Awards in London, England. The coveted award was bestowed on Slyce primarily because of their work with a world-renowned brand. “Tommy Hilfiger came to Slyce about a year and a half ago with a very ambitious vision: they wanted to build an app they could use at fashion shows and be able to shop the runway,” says Mann. Fashion show fanatics could shop their favourite looks from the show the minute they walk by.
The “snap to shop” app, TommyNow, was an instant success at London Fashion Week and over the course of several more fashion shows, Slyce expanded the app to allow users to purchase what other people have liked and shared and added an augmented reality feature allowing users to virtually try on clothing before buying it.
Slyce Canada’s headquarters in New Waterford, Nova Scotia is the epicentre of the company’s research and development activity. Most of the 30 employees there are deep learning engineers (a subset of AI, machine learning gives computers the ability to learn) who build the models that support automated detection. Slyce also has employees building training data to aid this mainframe schooling. “Any AI system is really only as good as the data that you train it on, so we’ve gone to great lengths to customize training data for each of our retailers,” says Mann. Slyce trains its AI with images from a client’s catalogue but is also programmed with similar images from scraping the Internet. “It’s a lot of brute force creation of this training data to get the models built up.”
What’s the next Slyce? Furthering their international footprint in Europe and the Middle East, which means training the AI in multiple languages like French and Arabic.