# Personal Knowledge Graphs in AI RAG-powered Applications with libSQL

DevFeed: [Personal Knowledge Graphs in AI RAG-powered Applications with libSQL](<https://devfeed.tech/articles/personal-knowledge-graphs-in-ai-rag-powered-applications-with-libsql-6020.md>)

Original publisher: [Read original article](<https://turso.tech/blog/personal-knowledge-graphs-in-ai-rag-powered-applications-with-libsql>)

Author: Volodymyr Pavlyshyn

Published: 2024-09-30T00:00:00Z

Content type: article

Language: en

Sources: [Turso Blog](<https://devfeed.tech/sources/turso-blog.md>)

Topics: [libSQL](<https://devfeed.tech/topics/libsql.md>), [Retrieval Augmented Generation (RAG)](<https://devfeed.tech/topics/retrieval-augmented-generation-rag.md>), [Graphs](<https://devfeed.tech/topics/graphs.md>), [Embeddings](<https://devfeed.tech/topics/embeddings.md>), [Local-First](<https://devfeed.tech/topics/local-first.md>), [Mobile](<https://devfeed.tech/topics/mobile.md>), [SQLite](<https://devfeed.tech/topics/sqlite.md>), [Databases](<https://devfeed.tech/topics/databases.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [databases](<https://devfeed.tech/tags/databases.md>), [embeddings](<https://devfeed.tech/tags/embeddings.md>), [graphs](<https://devfeed.tech/tags/graphs.md>), [libsql](<https://devfeed.tech/tags/libsql.md>), [local-first](<https://devfeed.tech/tags/local-first.md>), [mobile](<https://devfeed.tech/tags/mobile.md>), [rag](<https://devfeed.tech/tags/rag.md>), [sqlite](<https://devfeed.tech/tags/sqlite.md>)

## AI overview

The article explains how Kin uses libSQL in a local-first mobile AI application to store personal knowledge graphs and vector embeddings in an embeddable SQLite-compatible database. It describes the need to combine graph and vector capabilities, avoid synchronization costs across separate storage systems, and retain compatibility while adding vector search for RAG.

## Source excerpt

How Kin uses libSQL on mobile to store personal knowledge graphs alongside vector embeddings, powering local-first RAG with one embeddable SQLite file.