Cost and speed
Distilling Long-Tail User Behavior into Scalable Embeddings for Job Search
Our short summary of Indeed’s engineering write-up. The full story, with all the detail, is on their blog.
Read the full write-up on engineering.indeedblog.com (opens in a new tab)Problem
Online models needed fast, cheap scoring for millions of users, but could not use large sequence models.
Solution
They trained a heavy sequence model offline on full user histories, then distilled its knowledge into compact user embeddings that online models can use efficiently.
Want the architecture, code and numbers? They’re in the original post.
Read the full write-up on engineering.indeedblog.com (opens in a new tab)
Elastic