George He
① 他说过的话
6 条
其中最重要的一点是,代码已经在磁盘上、在本地了,尤其是如果你的智能体有上下文的话,那为什么要用一个预检索系统把它复杂化呢?
指向原始笔记的链接
And the largest part there is code is already on disk and local, especially if your agent has context, so why complicate it with a pre-retrieval system?
—— George He · [03:49]
你没法真正对一个公司数据集的文件夹做 grep,因为它不像代码那样组织有序。
指向原始笔记的链接
You can’t really grep the folders across a company dataset because it’s not organized like code.
—— George He · [04:13]
但这里的主要结论其实是,它最终归结为你所拥有的数据的规模,以及你愿意投入多少整理成本来维护那套合理的检索系统或数据管理系统。
指向原始笔记的链接
But the major takeaway here is actually it really comes down to the scale of the data that you have and how much curation effort you’re going to put in in order to maintain that retrieval system or that data management system that makes sense, right?
—— George He · [05:18]
一开始可能没问题,但如果你的 LLM 不理解位置和空间推理,到了某个时候就会崩掉。
指向原始笔记的链接
It might be fine at the start, but if your LLM doesn’t understand the location and spatial reasoning, it falls apart at some point.
—— George He · [14:58]
这就是为什么 Anthropic 和许多其他公司差不多已经放弃了尝试预索引数据,因为如果你的数据复杂,数据新鲜度真的很难实现。
指向原始笔记的链接
This is kind of why Anthropic, a lot of other companies have kind of given up on trying to pre-index data because data freshness is really hard to achieve if your datas are complex.
—— George He · [15:40]
但这里的要点和收获是:搜索帮你找到信息,而 harness 给你可信赖的结果。
指向原始笔记的链接
But the idea here and the takeaway is that search helps you find information, and the harness gives you the results that you can trust.
—— George He · [22:49]
② 出现在这些集
1 集
- 《搜代码不用向量库,搜公司文档为什么不行?》 — 作为嘉宾
③ 他谈到的
点进去有真内容 —— 本页主要出口
LlamaIndex · Claude Code · Anthropic · LlamaParse · LightParse · TurboPuffer · 智能体 · 预索引 · RAG · 向量数据库
