Abstract. Knowledge Management (KM) asks for information-intensive services over large amount of data, modeling the intellectual capital of an organization. To combine the expressiveness of logic-based languages with efficient information processing, we adopt a Knowledge Compilation approach to the extraction of "Core Competence" of a given company, a typical KM problem. In particular, we translate into a relational database schema the full logical description formalized in a Knowledge Base (KB), modeling organizational intellectual capital according to the formalism of Description Logics (DLs). Core Competence extraction is consequently performed through standard-SQL queries, while retaining the expressiveness of the logical representation. The service has been embedded in a system for Human Resource Management, I.M.P.A.K.T.3 , to show how Core Competence extraction performance significantly improves w.r.t. implementations exploiting DL reasoning engines.