Building a personalised home feed using Kafka Streams and Elasticsearch

06/12/2018 - 12:40 to 13:00
Palais Atelier
short talk (20 min)

Session abstract: 

At eBay Kleinanzeigen, we aim to inspire our users with a feed of the best items tailored to them. This becomes an interesting problem with more than 20 million monthly users and over 28 million live ads, with thousands of interactions taking place on our platform every second.

Some of the challenges that pop up are how to deal with new visitors, or ones that only visit occasionally. The posted items are often also very short-lived, as many get sold quickly. This requires us to be responsive (near real-time) with respect to our inventory and the users’ behaviour to help them find a match and be successful.

Technologies such as Kafka Streams and Elasticsearch allow us to approach the problem in a modern, elegant and scalable way, without the need for specialised clusters and long-running overnight batch jobs.

This talk is presented by eBay Tech.