Member of Technical Staff, ML Systems

Digitailor; Menlo Park (CA); On-site; USD 180,000 - 230,000; Full time

Job summary

Digitailor is seeking an experienced ML systems engineer to join its AI systems team in Menlo Park, CA. You will optimize training and inference performance for image, video, and world-model workloads, focusing on GPU kernels, distributed execution, and high-performance serving systems.

The role requires hands-on work with CUDA, Triton, and Nsight, plus collaboration across hardware, runtime, kernel, and model design to push performance gains in production.

Qualifications

• At least 2 years of experience in deep learning training or inference systems.
• Experience optimizing inference or training performance through GPU kernels, runtime systems, or distributed execution.
• Direct experience writing kernels for NVIDIA GPUs using CUDA, CUTLASS, Triton, or PTX.
• Authored core features in a training or inference framework (e.g., vLLM, SGLang, TensorRT-LLM, Megatron).
• Strong machine learning and CS fundamentals; degree in CS or a related quantitative field.
• Experience with PyTorch and Nsight; workloads like diffusion, video, image, or multimodal models.

Responsibilities

• Optimize training and inference performance for image, video, and world-model workloads.
• Profile bottlenecks at kernel, memory, system, and cluster levels using Nsight and related tools.
• Implement low-level GPU optimizations with 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, or RoCE.
• Build benchmarking and regression tools to maintain performance gains in production.
• Coordinate hardware, runtime, kernel, and model design for efficient execution.

Skills

Deep learning

GPU performance

Distributed systems

PyTorch

CUDA kernels

ML fundamentals

Education

Bachelor's degree in CS or related field

Tools

Nsight

CUDA

Triton

PTX

NCCL

RDMA

InfiniBand

RoCE

Job description

About The Role

Join an AI systems team building high-performance training and inference infrastructure for image, video, and world-model workloads. You will improve speed and efficiency across GPU kernels, distributed execution, and serving systems.

What You'Ll Do

Optimize training and inference performance for image, video, and world-model workloads. Profile bottlenecks at kernel, memory, system, and cluster levels using Nsight and related tools. Implement low-level GPU optimizations with 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, or RoCE. Build benchmarking and regression tools to maintain performance gains in production. Coordinate hardware, runtime, kernel, and model design for efficient execution.

What We'Re Looking For

At least 2 years of experience in deep learning training or inference systems, or distributed systems, including hands-on ML systems or GPU performance work. Experience optimizing inference or training performance through GPU kernels, runtime systems, or distributed execution. Direct experience writing kernels for NVIDIA GPUs using CUDA, CUTLASS, Triton, or PTX. Authored core features in a training or inference framework, such as vLLM, SGLang, TensorRT-LLM, or Megatron, rather than focusing only on deployment or integration. Strong machine learning and computer science fundamentals, with a degree in computer science or a related quantitative field. Experience with PyTorch, Nsight, and workloads such as diffusion, video, image, or other multimodal models is relevant. Compensation & Benefits Base

Salary

$180,000 to $230,000 annually. Visa sponsorship is available.

Location

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