Time series classi cation is a fundamental data science problem, providing understanding of dynamic processes as they evolve over time. The recent introduction of ensemble techniques has revolutionised this eld, greatly increasing accuracy, but at a cost of increasing already burdensome computational overheads. I present new time series classi cation technologies that achieve the same accuracy as recent state-of-theart developments, but with many orders of magnitude greater e ciency and scalability. These make time series classi cation feasible at hitherto unattainable scale.