Member of Technical Staff, ML Systems

About the Role

Own performance across the training and inference stack for image, video, and world-model workloads, from GPU kernels to distributed systems. You will work with an engineering and research team to make these workloads faster and more efficient.

What You'll Do

• Optimize training and inference performance across GPU kernels, memory, systems, and clusters.

• Profile workloads with Nsight and related tools, identify bottlenecks, and implement low-level optimizations in CUDA or Triton.

• Design and improve distributed training and inference engines for diffusion models.

• Improve communication across GPUs and nodes using technologies such as NCCL, RDMA, InfiniBand, or RoCE.

• Build benchmarks and regression tests to ensure performance improvements hold in production.

• Collaborate on hardware-aware kernel, runtime, and model design.

What We're Looking For

• At least 2 years of experience in deep learning training or inference systems, or distributed systems, including at least 1 year of hands-on ML systems or GPU performance work.

• Experience authoring core features in an inference or training framework such as vLLM, SGLang, TensorRT-LLM, or Megatron, rather than focusing only on deployment or integration.

• Direct GPU kernel optimization experience on NVIDIA GPUs using CUDA, Triton, CUTLASS, or PTX.

• Experience with PyTorch, Nsight, and multi-GPU or multi-node communication.

• Strong machine learning and computer science fundamentals, and a degree in computer science or a related quantitative field.

• Experience optimizing diffusion, video, image, or other multimodal workloads is valuable.

Compensation & Benefits

Annual salary range: $180,000 to $230,000 USD. Visa sponsorship is available.

Location

On-site in Menlo Park, California, United States.