A leading technology company is seeking a Principal / Senior GPU Software Performance Engineer in San Jose, CA. This role involves optimizing GPU workloads for training performance, enhancing throughput, and resolving bottlenecks in distributed systems. The ideal candidate will have strong skills in GPU performance engineering and experience with deep learning frameworks, particularly PyTorch. This position offers a chance to work in a collaborative environment focused on innovation and inclusivity.
Qualifications
Proven experience in GPU performance engineering for deep learning.
Strong practical experience with PyTorch and distributed systems.
B.S./M.S./Ph.D. in Computer Science, Computer Engineering, or Electrical Engineering.
Responsibilities
Lead performance for finetuning and RL training solutions on AMD GPUs.
Optimize multi-GPU/multi-node training and communication patterns.
Profile, diagnose, and resolve bottlenecks using standard tooling.
Skills
GPU performance engineering
PyTorch
Python
C++
distributed systems
Education
B.S./M.S./Ph.D. in Computer Science or equivalent
Tools
ROCm
Triton
Job description
A leading technology company is seeking a Principal / Senior GPU Software Performance Engineer in San Jose, CA. This role involves optimizing GPU workloads for training performance, enhancing throughput, and resolving bottlenecks in distributed systems. The ideal candidate will have strong skills in GPU performance engineering and experience with deep learning frameworks, particularly PyTorch. This position offers a chance to work in a collaborative environment focused on innovation and inclusivity.