2026-02-06-tiktok-algorithim-alternative

Monolith: Real Time Recommendation System With Collisionless Embedding Table (via HN)

Online training stage. After a model is deployed to online serving, the training does not stop but enters the online training stage. Instead of reading mini-batch examples from the storage, a training worker consumes realtime data on-the-fly and updates the training PS. The training PS periodically synochronizes its parameters to the serving PS, which will take effect on the user side immediately. This enables our model to interactively adapt itself according to a user’s feedback in realtime.