Nearly every machine learning model requires that the user specify certain parameters before training begins, aka "hyper-parameters". Finding the optimal set of hyper-parameters is often a time- and resource-consuming process. A recent breakthrough hyper-parameter optimization algorithm, Hyperband, can find high performing hyper-parameters with minimal training and has theoretical backing. This talk will provide intuition for Hyperband, explain it's use and why it's well-suited for Dask, a Python library that scales Python to larger datasets and more computational resources. Experiments find high performing hyper-parameters more quickly in the presence of parallel computational resources and with a deep learning model.