We present a comprehensive analysis of teaching assistant (TA) application preferences, with the goal of identifying whether there are significant differences in the courses students prefer to TA for based on student identity. Our data was gathered over the span of four years and represents 15,000+ individual applications for all courses in a computing department. Focusing on the dimensions of applicant program level (undergraduate versus Master’s students), gender, and international versus domestic student designation, we perform an analysis of application patterns. Our results show that program level, gender, and international status all play roles in student application behavior. Further, we identify specific courses that are preferred by various subsets of students, such as a strong affinity of students who have gone through a Master’s bridge program to apply to those courses and a tendency for women and non-binary students to apply at greater rates to our first-year seminar, data science, and databases courses. Finally, we investigate the relationship between instructor gender and applicant gender–revealing that when women and non-binary applicants prefer certain courses, these courses tend to also have a greater-than-average representation of women and non-binary instructors. As a result of this analysis, we present three recommended next steps to gain deeper understanding of the observed patterns. This work demonstrates that with a centralized TA application system, an institution can gain a comprehensive understanding of application behaviors–leading to informed decisions about where to potentially intervene, especially with an eye toward broadening participation in computing at all levels.