Source Graph
The owned map of systems, schemas, documents, permissions, events, and business meaning that agents use to act safely.
Production AI is not a prompt. It is a system of context, tools, permissions, traces, evals, and feedback loops.
Definition
A source graph is the navigable inventory of an organization's operational systems — schemas, documents, code, tickets, events, ownership, and permissions — connected by the relationships an agent needs to retrieve, cite, and act. It differs from a knowledge graph (which models domain concepts) and from a metadata or lineage graph (which models data movement) by being grounded in operational sources and shaped for agent use: every node has an owner, an access boundary, and a path back to the system of record.
Why it matters
Agents fail when they retrieve fragments without knowing where those fragments came from, who owns them, or how they relate to operational systems. A source graph gives retrieval a durable map.
- System and schema inventory
- Document and code ownership
- Permission-aware retrieval boundaries
- Lineage between events, data, and decisions
What we build
We connect warehouse models, operational APIs, documents, tickets, and code into a navigable substrate that can feed search, workflows, evals, and agent memory.
Related resources
Contracts, validation, lineage, freshness, and ownership for data that agents can safely use.
Retrieval pipelines that combine chunking, embeddings, metadata, reranking, permissions, and evaluation.
Turning support, Slack, sales, and product conversations into structured signal, knowledge updates, and eval cases.