A knowledge graph is a structured representation of entities and the relationships between them, typically stored as triples (subject — predicate — object) and queryable as a graph.
Why it matters
Google's knowledge graph (powering rich result panels) is the most visible example. Internal knowledge graphs at companies — Microsoft's Satori, LinkedIn's economic graph, Bloomberg's BBKG — power semantic search, recommendation, and analytics across structured + unstructured data.
Knowledge graphs complement vector embeddings: graphs encode discrete relationships ('Anthropic was founded by Dario Amodei'), embeddings encode fuzzy semantic proximity. Modern systems use both, typically extracting graph structure from text via LLM-based entity-and-relation extraction.
How Pith relates
Pith maintains a lightweight per-workspace knowledge graph: entities extracted from bookmarks, relationships between them, links from entities to wiki pages, briefings, and clients. The graph is exposed as the entities surface in the app.
See also
Last reviewed: 10 May 2026 · Licensed CC BY 4.0 · cite freely with attribution to Pith.