In this article, we discussed how to use Apache TinkerPop's time-agnostic traversal language with DSE Graph's time-prescient indexing system to efficiently query a graph database that models the world over time. We introduced the concept of "time machine" graphs and explained how to define a schema for such a graph, including global graph indices and local vertex-centric indices. We then demonstrated how to use TinkerPop's SubgraphStrategy to compile traversals into time-prescient forms that can be efficiently evaluated against DSE Graph's storage engine using its unique indexing system. We also discussed the importance of maintaining proper timestamps when modifying a graph and provided examples of how to do so in TinkerPop traversals. Finally, we shared some customer use cases for time-based graphs and acknowledged the contributions of various individuals who helped with this article. The future is yours!