The objective of the proposed research is to improve understanding of modern enterprise networks by addressing key limitations in traditional network telemetry. Preliminary analysis has shown that relying on traditional network telemetry like DNS monitoring alone causes blind spots that can affect the security posture of a network, and that attributing unexplained network activity requires reasoning over sparse and heterogeneous signals that does not scale with existing workflows. Building on these findings, this research develops methods to augment network telemetry with additional contextual data and automated reasoning using LLMs to attribute unexplained network activity. Finally, the work investigates whether such augmented analysis can extract higher-order organizational insights, specifically identifying critical organizational dependencies from network communication patterns.