Machine Learning Engineer - Model Compression Specialist

We are looking for a talented machine learning engineer to help us reduce the size of our EleutherAI/gpt-neox-20b and EleutherAI/gpt-j-6B and similar models using an MP method. The successful candidate will be responsible for designing and implementing an MP module to reduce the RAM requirements of these models, while maintaining their performance. Responsibilities: Develop and implement an MP module to reduce the RAM requirements of the EleutherAI/gpt-neox-20b and EleutherAI/gpt-j-6B models Work closely with the research team to ensure that the MP module does not significantly impact the models' performance Optimize the MP module to achieve maximum reduction in RAM usage while maintaining performance Document the MP module design and implementation for future reference Requirements: A degree in computer science, machine learning, or a related field Strong experience in machine learning model compression techniques, particularly MP methods Experience working with large language models such as EleutherAI/gpt-neox-20b and EleutherAI/gpt-j-6B Strong programming skills in Python and experience with deep learning frameworks such as PyTorch and TensorFlow Good communication skills and the ability to work collaboratively in a team environment