Cutting Edge Machine Learning Research & Development Programming Project

We are looking for an experienced machine learning developer to build some code and iterate the design of the code together with our team. The goal of the code is to construct a large scale echo state machine / reservoir computing model to perform inference on the SQUAD 1.1 data set at high accuracy.

A pytorch implementation would be ideal. We would recommend starting with the echotorch extension to pytorch, but this is not mandatory.

Available to work with is existing code packaged as a reference implementation of a major machine learning benchmark. Finished deliverable must be able to be submitted back to the benchmark organization for evaluation.

We are trying to understand how to do this in the most performant way possible.

Beyond this, we are interested in implementing an evolutionary algorithm to tune the size & weights of the echo state network.

Ideally you have experience doing work in NLP and at least a Masters in Data Science or Machine Learning. Experience working with the SQUAD 1.1 dataset would also be great. It would be very helpful if you are familiar with container technologies like Docker.