Infrastructure

Infrastructure is the durable substrate agents lean on: Vector DBs for similarity search and Knowledge Graphs for typed relationships. These systems are not the agent. They keep embeddings, entities, and edges available so RAG, memory, and company-model surfaces do not reinvent storage.

Pair with Agentic for the loops that read/write these stores, and with LLMs when serving produces the embeddings or completions that fill them. Prefer this layer when the bottleneck is persistence and query of structured or vector state — not tool-calling DX.

Members such as Qdrant, Milvus, and Neo4j show the two poles: ANN services and property graphs. Features like Knowledge Grounding & RAG and Persistent company model ≠ agent memory explain why the substrate exists in an OS: grounded retrieval and typed company structure.

Inline members for audit: Qdrant Milvus Neo4j. Features: Knowledge Grounding & RAG Persistent company model ≠ agent memory. Related terms referenced: Agentic, LLMs.

For this research catalog the acceptance test is boring and strict: after wiring the pack, can a second person replay the same scenario and get the same measurable outcome? If not, the page is still a brochure. Document the scenario, the expected number, and the log line that proves it. That habit is what separates a deepened taxonomy cell from a landscape screenshot.

Also record what you refused: tools not connected, corpora not ingested, metrics not collected. Refusals keep the cell honest when marketing pages claim universality. A taxonomy essay that only lists winners teaches less than one that names the skip conditions next to the install path.

Keep the essay technical enough that an engineer can argue install order — and short enough that a reader finishes it before opening the right-rail member list. Prefer primary docs and repo evidence over landscape blurbs when the right-rail packs disagree.