Build knowledge agents without embeddings (opens in new tab)

**Published:** March 19, 2026 | **Authors:** Ben Sabic, Hugo Richard | **Category:** Engineering *Deploy a file\-system\-based agent with Sandbox and Chat SDK* --- Most knowledge agents start the same way. You pick a vector database, then build a chunking pipeline. You choose an embedding model, then tune retrieval parameters. Weeks later, your agent answers a question incorrectly, and you have no idea which chunk it retrieved or why that chunk scored highest. We kept seeing this pattern...