---
title: "Comment by Kartik N V J K on \"Agent memory needs more than vector search\" | Hashnode"
description: "The line about two posts arguing opposite points scoring nearly identical in embedding space is the real failure, and it is why cosine similarity alone kept sur"
url: "https://hashnode.com/posts/agent-memory-needs-more-than-vector-search/6abbcc77cb659b470c8d2d80/comment/6abc15d3747881cc76719e4e"
---

[Discussion](https://hashnode.com/posts/agent-memory-needs-more-than-vector-search/6abbcc77cb659b470c8d2d80)

Comment [Agent memory needs more than vector search](https://hashnode.com/posts/agent-memory-needs-more-than-vector-search/6abbcc77cb659b470c8d2d80)

[Kartik N V J K](https://hashnode.com/@kartik-nvj)

[Sep 29](https://hashnode.com/posts/agent-memory-needs-more-than-vector-search/6abbcc77cb659b470c8d2d80/comment/6abc15d3747881cc76719e4e)

The line about two posts arguing opposite points scoring nearly identical in embedding space is the real failure, and it is why cosine similarity alone kept surfacing my stale notes. A reranker fixed relevance for me too, but the per-candidate latency forced me to cap the candidate set first. What candidate count did the BGE cross-encoder stay worth it at before the latency hurt the response?
