How to run your ML model Predictions 50 times faster?

With the advent of so many computing and serving frameworks, it is getting stressful day by day for the developers to put a model into production . If the question of what model performs best on my data was not enough, now the question is what framework to choose for serving a model trained with Sklearn or LightGBM or PyTorch . And new frameworks are being added as each day passes.

So is it imperative for a Data Scientist to learn a different framework because a Data Engineer is comfortable with that, or conversely, does a Data Engineer need to learn a new platform that the Data Scientist favors?

Add to that the factor of speed and performance that these various frameworks offer, and the question suddenly becomes even more complicated.

So, I was pleasantly surprised when I came across the Hummingbird project on Github recently, which aims to answer this question or at least takes a positive step in the right direction.