The popular online handmade items store Etsy has a problem. Its catalog contains over 40 million items, and it has 26.1 million active buyers. As a seller of highly customized goods, the product list can only grow. But searching for products in this vast catalog, and matching buyers to sellers poses some extraordinary challenges. Etsy’s answer? A machine learning acquisition. The company has just acquired AI search startup Blackbird Technologies to perform this critical function (Etsy Acquires Blackbird Technologies to Enhance Search Capabilities).
Blackbird is a deep-learning company that uses a GPU-based approach to search, with a Theano deep learning framework. According to Etsy’s release, Blackbird adds:
- Machine learning that analyzes user behavior, unstructured data, and other variables to suggest relevant and personalized search recommendations;
- Natural language processing to understand complex search queries;
- Deep Learning-based image recognition techniques to index and catalog photos, which help provide more relevant search results; and
- Spelling correction and predictive type-ahead to make searching faster and more intuitive.
- Blackbird’s talented employees, including its co-founders, who serve as the company’s CEO and CTO, will join Etsy. The team possesses a deep expertise in Artificial Intelligence, search, and distributed systems and has direct experience working in these areas for some of the largest technology companies in the world.
The addition of experienced analytics employees who are versed in deep learning is a notable plus for Etsy’s future. This is likely to become increasingly important as the company grows in the area of 3d Printing. 3d-printed goods are now making a strong appearance on the platform, and are capable of expanding the “handmade” catalog geometrically.
In any case, AI-based search is a good addition, and helps Etsy with an increasingly competitive environment.