Domain knowledge extraction
person concept tool org talk claim — click a node to jump to its page; hover an arrow for the relation
Domain knowledge extraction is the process of capturing a company's institutional expertise, processes, and know-how and encoding it as structured context or defined skill sets that can be supplied to AI systems.
Core concept
Tom Blomfield advocates for domain knowledge extraction as a foundational step in building AI-augmented organizations. He frames it as "extracting the domain knowledge from your company and defining it as a as like context or a set of skills" — the idea being that tacit organizational knowledge must be made explicit and machine-readable before AI agents can usefully act on it. 1
Significance
The concept sits at the intersection of knowledge management and AI engineering: rather than expecting general-purpose models to infer company-specific norms, workflows, or domain expertise from scratch, domain knowledge extraction externalizes that expertise into context or skill definitions that can be reliably injected into agent pipelines. This makes organizational intelligence transferable, auditable, and reusable across AI-powered workflows.