Kuba Rogut distinguishes retrieval augmented generation from one-shot vector search and describes agentic retrieval as iterative searchingRetrieval is the process of selecting relevant stored information and returning it to an AI system for the current task. and reasoning across multiple tools.
He discusses Cursor’s codebase indexing and reported internal evaluation resultsEvaluation is the systematic assessment of a system's behaviour or results against explicit criteria., then contrasts repeated local searches with indexing costs that can be reused across sessions. He argues that selecting useful context remains important even with large context windows.
Watch on YouTube
