I describe a research roadmap for going beyond the state of the art in AI, and Qualitative Spatial and Temporal Reasoning (QSTR) in particular, and building hybrid architectures for AI that involve also robust and dynamic symbolic computation. Simply put, QSTR is a major field of study in AI that abstracts from numerical quantities of space and time by using qualitative descriptions instead (e.g., precedes, contains, is left of), with applications in a plethora of areas and domains such as smart environments, intelligent vehicles, and unmanned aircraft systems. Ultimately, I want to push the envelope in AI by defining tools for tackling dynamic variants of fundamental spatio- temporal reasoning problems, i.e., spatio-temporal problems stated in terms of changing input data, and integrating these tools into the bigger context of highly active areas such as neuro-symbolic AI.