Description:
JOB LEVEL P50 ADDITIONAL JOB LEVELS P55 - EMPLOYEE ROLE Individual Contributor What you will be working on: Optimize PyTorch-based training code for large scale distributed training Enhance existing training frameworks to better accommodate FP8 and mixed precision Ensure efficient utilization of GPU resources for large scale distributed training Ensure efficient setup and utilization of network for large scale distributed training Quality and performance analysis between data types such as
Apr 18, 2024;
from:
dice.com