Member of Technical Staff, ML Systems

True; Menlo Park (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

The company is seeking an experienced ML systems engineer to optimize training and inference workloads for image, video, diffusion, and world-model tasks at their Menlo Park, CA site. You will profile bottlenecks, implement low-level GPU optimizations with CUDA and Triton, and help design distributed engines for diffusion models.

This role requires 2+ years in ML systems, strong CS fundamentals, and hands-on experience with PyTorch, Nsight, NCCL, and RDMA.

Qualifications

• 2+ years in deep learning training or inference systems with ML systems or GPU performance experience.
• Experience authoring core features in an inference or training framework (e.g., vLLM, SGLang, TensorRT-LLM, Megatron).
• Hands-on GPU kernel development for NVIDIA hardware using CUDA, CUTLASS, Triton, or PTX.
• Strong machine learning and CS fundamentals, usually with a CS-related degree.
• Experience with PyTorch, Nsight, NCCL, and RDMA.
• Experience optimizing inference or training performance for diffusion, video, image, or multimodal workloads.

Responsibilities

• Optimize training and inference performance for image, video, diffusion, and world-model workloads.
• Profile bottlenecks across kernels, memory, systems, and clusters using Nsight and related tools.
• Implement GPU optimizations using CUDA and Triton.
• Design and improve distributed training and inference engines for diffusion models.
• Improve communication performance across GPUs and nodes using NCCL, RDMA, InfiniBand, and RoCE.
• Build benchmarks and regression tests to ensure performance improvements hold in production.
• Contribute to hardware-aware co-design across kernels, runtimes, and models.

Skills

ML systems

GPU performance

CUDA

Triton

PyTorch

Distributed systems

Education

Bachelor's degree in CS or related field

Tools

Nsight

NCCL

RDMA

InfiniBand

RoCE

CUDA

CUTLASS

Triton

Job description

ABOUT THE ROLE

Join an AI and machine learning team building efficient training and inference systems for image, video, and world-model workloads. You will improve performance across GPU kernels, distributed execution, and multi-node serving, helping make demanding model workloads faster and more efficient.

WHAT YOU'LL DO

• Optimize training and inference performance for image, video, diffusion, and world-model workloads.
• Profile bottlenecks across kernels, memory, systems, and clusters using Nsight and related tools.
• Implement low-level GPU optimizations using CUDA and Triton.
• Design and improve distributed training and inference engines for diffusion models.
• Improve communication performance across GPUs and nodes using technologies such as NCCL, RDMA, InfiniBand, and RoCE.
• Build benchmarks and regression tests to ensure performance improvements hold in production.
• Contribute to hardware-aware co-design across kernels, runtimes, and models.

WHAT WE'RE LOOKING FOR

• At least 2 years of experience in deep learning training or inference systems, distributed systems, or related work, with hands-on ML systems or GPU performance experience.
• Experience authoring core features in an inference or training framework, such as vLLM, SGLang, TensorRT-LLM, Megatron, or a similar system.
• Hands-on experience writing GPU kernels for NVIDIA hardware using CUDA, CUTLASS, Triton, or PTX.
• Strong machine learning and computer science fundamentals, typically supported by a degree in computer science or a related quantitative field.
• Experience with PyTorch, Nsight, NCCL, and RDMA is relevant.
• Background optimizing inference or training performance, ideally for diffusion, video, image, or other multimodal workloads.

COMPENSATION & BENEFITS

• Base salary: $180,000 to $260,000 annually.
• Visa sponsorship is available.

LOCATION

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