Senior/Staff Software Engineer, ML Training Optimization

What you’d do

Lead design and operation of training performance optimizations such as parallelism, mixed precision, and GPU kernels to scale foundation model training, and own the roadmap for RL framework performance. Mentor engineers and drive profiling of throughput bottlenecks across compute, compiler, framework, and networking. Found on the employer's careers page on Oct 17, 2026. Apply through the link.

You’ll need

• 7+ years total experience
• 4+ years on large-scale model training platforms
• Record of leading projects end to end
• Training performance optimization (parallelism, mixed precision, GPU kernels)
• Profiling and root-cause analysis of throughput bottlenecks