“The primary technical topic: using vector embeddings to improve code retrieval.”

“The method of finding relevant information using embeddings.”

“A valuable tool integrated into the Cursor agent harness, allowing agents to make natural language queries to find files. It's powered by Cursor's own embedding model.”

“semantic search that's part of that component”

“when semantic search versus lexical search like makes sense. You can do a reex search, you can do a semantic search, you can do both.”