Candidate should be able to:
Interact with deep learning researchers and experience with deep learning frameworks.
Transform computational graph representation of neural network model. Develop deep learning primitives in math libraries.
Implement various distributed algorithms such as model/data-parallel frameworks, parameter servers, data flow-based asynchronous data communication in deep learning frameworks.
Optimize code for various computing hardware backends.
Design, develop optimize for deep learning training and inference frameworks.
Conduct design and development to build and optimize deep learning software.

Candidate should have:
Excellent communication skills and interpersonal skills to be a team player.
ability to independently handle technical issues and come up w solutions
Ability to be a self-starter and highly proactive lead with a passion to work closely with customers in emerging areas applications