About the Role

In this role, you are responsible to build a highly scalable inference framework for our future chip generations. You will participate in fundamental architectural decisions and have the opportunity to contribute to upstream open-source projects.

What you will do:

• Research, develop, and maintain the machine learning inference framework used for SEMRON architectures.
• Collaborate with machine learning, compiler and hardware engineers/researchers to facilitate the quantization/optimization of NNs for SEMRON’s hardware.

What you should bring in:

• Proficiency with machine learning frameworks like Tensorflow, PyTorch in Linux environments including the ability to extend those framework with custom C/C++/CUDA code
• A deep understanding of machine learning algorithms and modeling techniques, including but not limited to semi-supervised or weakly supervised learning, transfer learning, quantization, optimization, and large language models
• Knowledge of various operations methodologies, including MLOps, ModelOps, and LMOps

Helpful but not required:

• Experience with State-of-the-art NN compression methods like Adaround, QDrop, QUIP, or GPTQ
• Experience with typical tools used in ML environments like HuggingFace’s transformers or DeepSpeed