Agents · Retrieval · Scale
How does an agent actually search a large, distributed set of documents? Classic RAG chunks documents, embeds the chunks, and retrieves the top-k nearest vectors. Each of those steps has a failure mode. Below, you can watch each one break — then see how an agent plans around it.
A document is split into fixed-size chunks, each chunk is embedded, and every chunk lands as a point in vector space. Drag the sliders and watch a key fact get sliced in half at small chunk sizes.
200 chunks from three source documents projected into space. Run a query: the nearest k chunks light up cyan; chunks that are truly relevant but semantically distant pulse coral — the near-misses top-k never sees.
“What changed in the Q3 refund policy and who approved it?” The policy lives in one store; the approval email lives in another. Watch both strategies attempt it.
Five structural failure modes that get worse as corpora grow.
Facts that straddle a chunk edge are split into two half-facts; neither embedding fully captures the claim, so neither ranks well.
If an answer needs 12 supporting chunks and k is 5, no amount of embedding quality saves you. Raising k adds noise and blows the context budget.
Embeddings freeze a snapshot. Edited or deleted documents keep answering questions until the next re-index run.
One query vector cannot express “find X, then follow its reference to Y.” Single-shot retrieval flattens a chain into one lookup.
Real corpora live in separate stores with separate indexes. A query against one silo is blind to relevant chunks in another.
Five quick questions. Instant feedback with explanations.