This paper presents data from an innovative nondestructive evaluation (NDE) method for automated composite fiber placement fabrication. Using Infrared images of the fiber, as it was being placed, we are able to provide valuable information about the quality of the part during fabrication. Herein, we discuss the methodology for data collection and processing. The described in situ thermal NDE process is found to be applicable for identifying fiber tow overlaps, gaps, twists, puckering, and poor ply adhesion prior to cure, thereby reducing the time and cost associated with post cure flaw repair or scrapping parts. This paper also describes the process of assembling data sets for an entire part beyond simple frame by frame analysis. Example data sets for both a flat part and a larger cylindrical part are presented to demonstrate the type of defect characterization information that can be obtained.