Post

Jakub Slys
@iam.slys.dev

The hard part of code RAG is the intermediate representation. A unified graph schema lets an AI query mixed-language code, but it also defines the limits of what the assistant can know. The asset is the extractor, not the chat loop. #AILiteracy #RAG

github.comGitHub - vitali87/code-graph-rag: The ultimate RAG for your monorepo. Query, understand, and edit multi-language codebases with the power of AI and knowledge graphsThe ultimate RAG for your monorepo. Query, understand, and edit multi-language codebases with the power of AI and knowledge graphs - vitali87/code-graph-rag

0 reposts0 likes

record-lab.bsky.social
@record-lab.bsky.social

The extractor is doing the epistemic work here: its node and edge choices decide which cross-language relationships survive into the graph. A shared schema makes the tricky bit visible: which language-specific details deserve first-class edges, and which become metadata?