Over multiple decades, NRL has focused on satellite data processing and algorithms and our input satellite data suite has grown from a single satellite to nearly 40, while our product suite has grown at an even faster pace. With the impending launch of NPP on October 25th, 2011 our data volume is set to jump from the current 300 GB/day to 3 TB/day and future launches such as JPSS and GOES-R will further increase this volume. As a consequence of past growth, NRL's satellite processing algorithms have grown organically, producing a large body of “spaghetti code” where many scripts are simply copied and pasted and then modified, sometimes by persons who were not the original creator. Small modifications to the processing environment require changes to a very large number of scripts, thus costing time and creating error opportunities. In order to improve maintainability, conciseness, upgradeability for new sensors, and ease of use for both programmers and non-programmers, we have developed an object oriented approach to satellite data processing through a mixture of Python and TeraScan. This scheme will automatically generate satellite imagery and derived products (cloud properties, ocean surface wind speed, aerosol optical depth, dust and volcanic dust detection, etc) covering many geographical regions of interest, both stationary (e.g. CONUS) and transient (e.g. hurricanes). A GUI interface will also be provided for adding new regions and adding new products to those regions. In addition this scheme should make the transition path to operations more clear and simple and assist us in handing growth due to new sensors.