What an Enterprise Context Layer Actually Is (Prukalpa)
Prukalpa (co-founder, Atlan) defines the enterprise context layer: “the system that turns knowledge, expertise, and norms into machine-usable context for AI, across the heterogeneous landscape of data, business systems, and AI tools.” Effectively the organizational-scale version of this wiki’s thesis — a compounding, governed knowledge substrate an LLM/agent reasons over.
Three types of context
- Knowledge — the semantic map of the business (entities, metrics, definitions) — i.e. a knowledge-graph of the org.
- Expertise — how work actually gets done (procedures, workflows, playbooks) — the organizational analog of agent-skills (cross-wiki: procedures as loadable capability).
- Norms — policies, permissions, compliance (acceptable action).
Explicitly broader than a semantic layer or data catalog (which serve analytics metrics / humans finding tables): a context layer spans data + semantics + skills and runs as an “operating system.”
Five capabilities
Context mining (extract hidden knowledge) · development lifecycle (versioning & governance of context assets) · learning loops (promote agent experience into durable org knowledge) · activation/retrieval (deliver context across interfaces) · governance & observability (prevent knowledge decay).
The compounding thesis — why it’s on-thesis here
“Organizations whose context compounds will win, because the tenth agent is dramatically smarter than the first when built on accumulated, governed knowledge.” This is precisely compound-engineering / the llm-wiki “compounding artifact” bet and gbrain‘s compounding brain — now stated as enterprise infrastructure (and gbrain already has a multi-user “company brain” thread, so this is the generalized, vendor-framed version). The “learning loops” are the org-scale cousin of self-improving-agents (cross-wiki). Its versioning/governance/observability emphasis is the maintenance discipline this wiki enforces by hand (lint/synthesis), made an explicit product requirement. Audience: enterprise CTOs/CIOs. (Caveat: vendor-founder framing.)
Related
compound-engineering · llm-wiki · gbrain · knowledge-graph · agent-skills · tools-for-thought · synthesis