Staff Software Engineer, Training - Remote

Overview

The Staff Software Engineer for Training will focus on optimizing the training stack, enhancing performance and efficiency in distributed systems and ML infrastructure.

In Short

• Drive down wall-clock time to convergence by profiling and eliminating bottlenecks.
• Design, build, and optimize distributed training systems using PyTorch.
• Implement efficient low-level code with CUDA and custom kernels.
• Optimize workloads for hardware efficiency including CPU/GPU balance.
• Develop monitoring and debugging tools for large-scale runs.

Requirements

• Deep experience in distributed systems, ML infrastructure, or high-performance computing (8+ years).
• Production-grade expertise in Python.
• Low-level performance mastery with CUDA/cuDNN/Triton.
• Experience with PyTorch and training jobs using various parallelism techniques.
• System-level mindset with a track record of tuning hardware–software interactions.

Benefits

• Work in a dynamic and innovative environment.
• Opportunity to contribute to cutting-edge ML technologies.
• Collaborate with a talented team of engineers and researchers.
• Flexible work arrangements including remote options.
• Competitive salary and benefits package.