Vespa at Zedge - providing personalization content to millions of iOS, Android and web users

Zedge, a company known for its popular app providing wallpapers, ringtones, and game recommendations, adopted Vespa, an open-source search and recommendation engine, to enhance its content discovery capabilities for millions of users across Android, iOS, and the web. Faced with increasing complexity in its existing systems, Zedge initiated a pilot project to evaluate Vespa's potential benefits, including its scalability, minimal operational requirements, and robust feature set, such as built-in tensor processing and support for Tensorflow models. The pilot was successful, leading to Vespa's integration into Zedge's architecture, resulting in a more dynamic and efficient recommendation system with reduced computational demands. Zedge also implemented continuous integration and monitoring solutions using tools like Jenkins, Docker Compose, Prometheus, and Grafana to ensure smooth deployment and performance tracking. The transition to Vespa has paved the way for further advancements in Zedge's search and recommendation capabilities, with potential enhancements through machine learning integrations.