We added language routing work for RAG queries. The goal is to keep knowledge access language-aware without forcing every path into translation first.
This matters for retrieval because evidence can exist in a different language from the user query. Language routing gives the system an explicit place to decide how to handle that case.
Datasource routing work was added so retrieval can target relevant sources instead of searching every corpus the same way.
We clarified the boundary between conversation memory, retrieved evidence, and answer synthesis in the Agentic RAG runtime.
Keyword retrieval work gave the RAG layer a lexical path for exact terms, identifiers, and phrases.