LHC experiments will be faced with large amounts of data to be processed and analyzed during a period of operation of about 15 years. Given the long timescale of the LHC project, it is of vital importance to make the analysis software transparent for the users and easily maintainable for the developers. For this reason all LHC experiments have decided to develop their software using object-oriented technologies.
This talk will aim at describing the current state of the object-oriented track reconstruction software in the Inner Tracker of the CMS experiment at CERN. Focus will be put on the design and structure of the code rather than on algorithmic details. If time permits, key performance figures from simulated benchmark channels will also be presented.