从零构建RAG(4)：Typesense向量数据库实战 | RAG with Typesense

转自 https://www.youtube.com/watch?v=adPi3a8fq4c
转自 https://www.youtube.com/watch?v=adPi3a8fq4c

本视频是"从零构建RAG流水线"系列的第四集，由AI工程师Krishna主讲，内容围绕如何将RAG管道中的向量数据迁移到云端展开。课程引入了 Typesense —— 一款开源的闪电级搜索引擎，并演示如何将其作为托管在云端的向量数据库使用。通过本课，你将学会完整的数据摄入流程：从文本分块、向量嵌入到存入 Typesense，并利用其内置的向量相似性搜索实现 RAG 查询，适合希望从本地向量库迁移到生产级云端方案的学习者。

This is Episode 4 of the "Build RAG Pipeline From Scratch" series by AI engineer Krishna. Building on previous episodes, this video focuses on moving the RAG pipeline's vector storage to the cloud using Typesense — a lightning-fast, open-source search engine that supports vector search. You'll learn how to ingest chunked and embedded text into a Typesense cloud instance and perform RAG queries using its built-in vector similarity search. Great for learners ready to move beyond local vector stores to a production-ready cloud solution.

原版视频：https://www.youtube.com/watch?v=adPi3a8fq4c