ML Systems Engineer: Scalable Training & GPU Kernels

Motional; Boston (MA); Hybrid; USD 144,000 - 192,000; Full time

Benefits offered by this job

Health insurance

Dental insurance

Vision coverage

401k with company match

Health savings account

Life insurance

Pet insurance

Job summary

Motional AD Inc. is seeking a Machine Learning Systems Engineer to join the ML Acceleration team in Boston.

You will help researchers train frontier models at scale, focusing on speed, cost, reliability, and throughput. You will work at the intersection of machine learning research and high-performance systems engineering, crafting GPU kernels, optimizing data pipelines, and improving distributed training workflows to accelerate convergence for next-generation models.

Qualifications

• Bachelor’s, Master’s degree, or PhD in Computer Science, Computer Engineering, or a related technical discipline.
• Strong proficiency in Python.
• Extensive hands-on experience with PyTorch.
• Experience optimizing machine learning model execution during training and inference, alongside a strong understanding of fundamental machine learning concepts.
• Exceptional analytical and problem-solving skills with a data-driven approach.

Responsibilities

• Performance Profiling & Optimization: identify bottlenecks in data loading, gradient computation, and communication; implement kernel fusion, sharding, tiling.
• Distributed Training: optimize pipelines using PyTorch Distributed.
• Kernel Development: design and maintain high-performance GPU kernels in Triton or CUDA.
• Data Pipeline Engineering: optimize data loading pipelines to maximize training throughput.

Job description

Motional AD Inc. is seeking a Machine Learning Systems Engineer to join the ML Acceleration team in Boston.

You will help researchers train frontier models at scale, focusing on speed, cost, reliability, and throughput. You will work at the intersection of machine learning research and high-performance systems engineering, crafting GPU kernels, optimizing data pipelines, and improving distributed training workflows to accelerate convergence for next-generation models.