Social feed and video recommender for Frever
Social feed and video recommender for Frever
Frever’s unique video-creation and social content sharing app enables their users’ creativity. Users create personalized avatars and express themselves through music videos, stories, and vlogs.
Background
In Frever’s app, users create and share short animated videos, by creating avatars and building scenes, adding voice and music. The videos are shared and content is discovered through a TikTok-like feed. Users can share, comment, and like videos of fellow creators, as well as produce remixes. In order to offer the best possible user experience, a relevant and personalized feed, that captures users’ engagement and stimulates creativity is of central interest.
Solution
Engineers at Frever and Modulai teamed up in close collaboration to create an end-to-end machine-learning-based feed recommender system. A multi-model system architecture was developed and populates the feeds of every user. Information about the content of the video as well as indicators of the users’ preferences is taken into account to ensure the best possible experience and relevance.
Tools/Tech
A fully cloud-based solution was developed for the training, validation, and deployment of content-based and collaborative filtering models. Coding was mainly done in Python as dockerized microservices. A set of internal APIs passes input data and recommendations between the producing and consuming systems.