AISoftware

Recombee offers recommendation web services

By 16 March 2017 No Comments

Hundreds of websites rely on recommendation services. Whether it is an online shop, a job offer aggregator, video-on-demand or even a news portal, they all provide personalized articles or recommend items that the users are likely to buy. Recombee offers an intuitive API (Application Programming Interface) that allows companies to generate recommendations in real time.

The service can deliver more than 500 recommendations per second in the form of homepage personalization, similar items recommendation, personalized emailing campaigns or search results sorting. All recommendation models work in real time and take into account users’ actions such as views, purchases or bookmarks. The Recombee engine is fully customizable and is constantly being improved with the use of artificial intelligence.

Apart from the RESTful API, Recombee offers Software Development Kits for many different programming languages such as Java or Ruby. Additionally, it equips the companies with a graphical user interface for monitoring the Key Performance Indicators in real time. The whole engine is free of charge for up to 99,999 recommendation requests per month.

Recombee was launched in December 2015 by five founders. Pavel Kordik, the CEO, is a postdoc with 20 years of experience in artificial intelligence and machine learning. Tomas Rehorek, the CTO, has profound experience in the area of recommender systems. They were joined by Ondrej Fiedler, the current lead developer, Tomas Barton, a development and operations expert, and Antonin Kral, a serial entrepreneur who advises the team on further expansion.

Thanks to the founders’ academic background, they continue to do a lot of research in the area of recommender systems and machine learning. This way, the algorithms used by Recombee are constantly refined and improved. The quality of their product, along with the customers’ references, the company grows organically with almost no marketing. At the moment, it serves 100 requests per second; this number doubles every third month.

When asked about the plans for the nearest future, Pavel Kordik, the CEO, said: “Currently, we are able to provide recommender as a service to companies with dozens of millions of users and products with the plan to scale up. We work hard to make our service even more useful and profitable not only for large enterprises but also for small and medium-sized companies.”

Author Bogna Haponiuk

Engineer by education, writer by passion, creative spirit by nature, innovation enthusiast by choice.

More posts by Bogna Haponiuk

Leave a Reply