In November 2024, a paper titled "Unifying Generative and Dense Retrieval for Sequential Recommendation" was released. The paper introduced a new state-of-the-art model, LIGER, improving over a previous generative recommender model, TIGER (see our blogpost comparing TIGER and BaseModel), with particularly strong gains in cold-start settings. In this

In this article, we dive into how a gateway can handle complex SQL queries by streaming results in manageable chunks instead of trying to load gigabytes of data at once. What I really like here is the transition from a blocking, synchronous model to an asynchronous coroutine-based approach.

🚀 Introduction Recommender systems have undergone significant evolution since their inception, gaining momentum after the landmark research paper published in 1994. The first research paper on recommender systems was published by Resnick, Iacovou, Suchak, Bergstrom, and Riedl. This field has expanded considerably over the years, driven by key challenges and competitions

Overview In previous articles, we primarily focused on the gateway side, where we handle requests from OQS and fetch the necessary data from workers. The gateway operates as both a client and a server.  To manage the heavy workload of handling 2,160 workers with only 64 threads, we use