Agent Engineering Wiki

The retrieval stack is consolidating into single, self-hosted gateways rather than staying bespoke per project: Orbit packages file RAG, vector RAG across five-plus backends (Chroma, Qdrant, Pinecone, Weaviate, pgvector, FAISS), and natural-language-to-query translation over SQL, NoSQL, and REST sources into one open toolkit — treating "which store, which query language" as a routing decision inside the gateway rather than a separate integration per source.

Deterministic retrieval is a live alternative to embedding everything: a production Postgres pattern assembles context by writing a plain SQL query ("how would a human solve this?") instead of reaching for similarity search by default, reserving HNSW-indexed vector search — with quantization for roughly 4x faster lookups — for the genuinely fuzzy slice of the problem. It's the structured-recall argument agent memory already makes, applied to what an agent fetches rather than what it remembers.

Fetching itself is a grounding cost, not just a token-cost line item: a raw Wikipedia page runs roughly 68,240 tokens versus 3,000-5,000 once converted to markdown by a stealth-browser fetch tool — the same information, with most of the difference being boilerplate the model has to read before it can ground on the part that matters (see agent cost for the token-price side of the same fact).

Attribution is now a measured axis, not a vibe: ResearchQA scores whether an LLM's answer over scientific papers is actually backed by verifiable citations rather than just scoring the answer text, and a tool-adaptive reranker conditions its reranking on which retrieval tool produced each candidate — both targeting the specific failure mode where a model answers fluently past what its retrieved context actually supports.

This is the wiki's first pass at grounding as its own obstacle, split out from agent memory: a single-gateway consolidation of the retrieval stack (Orbit), a deterministic SQL-first alternative to embedding everything, a measured token cost for raw web fetches, and the first dedicated attribution benchmarks and rerankers scoring whether an answer is actually backed by what was retrieved.

Grounding is the trust layer underneath every agent answer that cites a source or claims a fact: get it wrong and the agent is fluent but unverifiable, which is worse than an obvious failure because users don't know to distrust it. The engineering job splits three ways — pick the retrieval architecture (vector, graph, SQL, or a gateway spanning all three), budget the token cost of fetching before it enters context (cross-ref cost), and measure attribution directly rather than assuming a fluent answer is a grounded one.