Every technology has an applicability range and a hype curve. While working with statistical-based deep learning ML, thinking of what should come next.