Responsibilities :

- Design and develop performance optimization techniques for machine learning (ML) models to run efficiently on Client hardware.

- Develop and maintain software that efficiently deploys ML models built on various ML frameworks (pytorch, tensorflow, caffe2, etc.) to Client hardware.

- Collaborate and communicate with the other teams that work on hardware level implementations and optimizations for the various ML operations/layers.

- Follow agile software development methodology, unit testing and continuous integration.

Requirements :

- Strong hands-on experience in training and inference of ML models using ML frameworks like Pytorch, Tensorflow, etc.

- Excellent programming skills in Python, C/C++

- Good Knowledge on neural network workings.

- Good Knowledge of mathematical computations/workings of the most commonly used neural network layers.

- Familiarity on the workings and basic architecture of various ML models related to computer vision, NLP, autonomous driving, etc.,

- Familiarity with neural network performance optimization techniques including but not limited to quantization, pruning, etc.,

- Familiarity with ONNX is a plus.

- Familiarity with SIMD Computer Architectures is a plus.