This paper describes a reverse engineering case study with an industrial-scale software system. The steps followed in the study involved: establishing a more comprehensive domain model for representing structural facts about C/C++ software; adapting a robust C/C++ parser that focuses on source code before the preprocessing stage; decomposing or clustering the extracted artifacts in various ways for different understanding needs; generating reports to highlight anomalies in the system; and clarifying knowledge about its conceptual architecture. Also summarized are lessons learned from this experience.