---
title: "Discussion on \"How I cached intention, not queries\" | Hashnode"
description: "Discussion on \"How I cached intention, not queries\". Cut LLM token usage 80-90% and latency from 40s to 10s by caching user intent with semantic embeddings and Qdrant, not exact queries"
url: "https://hashnode.com/posts/how-i-cached-intention-not-queries/6a1f886d1d44cf54bad48e57"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [How I cached intention, not queries](https://precious-o.hashnode.dev/how-i-cached-intention-not-queries?utm_source=hashnode&utm_medium=feed)

Cut LLM token usage 80-90% and latency from 40s to 10s by caching user intent with semantic embeddings and Qdrant, not exact queries

Jun 3 · 3 min read · precious-o.hashnode.dev  
Read the article

-   [#ai](https://hashnode.com/tag/ai)
-   [#ai-engineering](https://hashnode.com/tag/ai-engineering)
-   [#llm](https://hashnode.com/tag/llm)

## Written by

[Precious Obinna](https://hashnode.com/@Prehqsy)

@Prehqsy · Lagos Nigeria

  
[View profile](https://hashnode.com/@Prehqsy)

Developer

I'm a developer. I'm gonna rule the world soon. I'm currently on the path of Mastery. Valar Morghulis.

## Responses

Markdown supported · ⌘↵ to post

No responses yet.
