ML Systems Engineer: High-Performance Distributed Training

Motional; San Francisco (CA); Hybrid; USD 144,000 - 192,000; Full time

Benefits offered by this job

Medical

Dental

Vision

401k with company match

Health saving accounts

Life insurance

Pet insurance

Job summary

Motional is seeking a Machine Learning Systems Engineer to join our ML Acceleration team in San Francisco. You will help build core systems enabling researchers to train frontier models at scale, focusing on speed, cost, reliability, and throughput.

You will profile and optimize data loading, gradient computation, and communication, develop high-performance GPU kernels, and contribute to distributed training pipelines using PyTorch Distributed.

Qualifications

• Bachelor's, Master's, or PhD in CS/CE or related field.
• Strong Python proficiency.
• Experience with PyTorch and ML training/inference optimization.

Responsibilities

• Performance profiling and optimization of data loading, gradient computation, and communication.
• Distributed training pipeline optimization using PyTorch Distributed.
• Design and development of high-performance GPU kernels in CUDA/Triton.
• Data pipeline optimization for training throughput.

Skills

Python

Analytical thinking

Problem solving

Education

Bachelor's, Master's or PhD in CS/CE or related field

Tools

PyTorch

Nsight

CUDA

Triton

Job description

Motional is seeking a Machine Learning Systems Engineer to join our ML Acceleration team in San Francisco. You will help build core systems enabling researchers to train frontier models at scale, focusing on speed, cost, reliability, and throughput.

You will profile and optimize data loading, gradient computation, and communication, develop high-performance GPU kernels, and contribute to distributed training pipelines using PyTorch Distributed.