Staff Machine Learning Engineer - ML Training Infrastructure

GM•Sunnyvale, CA

2d•Remote

About The Position

We are seeking an experienced, technically strong, impact-driven expert in ML Training Infrastructure with a demonstrated ability to lead through hands-on technical work. In this role, you will be responsible for defining the technical direction and driving the design and development of scalable, reliable, and high-performance AI/ML platform infrastructure that enables advanced AI research and model development at scale. As a Staff ML Engineer, you will operate as a technical leader across initiatives, partnering closely with machine learning engineers, research scientists, and platform teams to shape architecture, drive major technical decisions, and deliver state-of-the-art AI infrastructure that enables the future of intelligent driving technologies across General Motors vehicles.

Requirements

Bachelor's degree or higher in Computer Science or a related field, or equivalent practical experience.
7+ years of professional software engineering experience.
5+ years of specialized experience in AI/ML infrastructure, such as enabling distributed training for large-scale ML models.
Strong programming skills in Python, with deep proficiency in frameworks such as PyTorch (preferred), TensorFlow, or similar ML systems.
Proven experience designing and operating distributed systems for ML training, including distributed computing, GPU computing, and cloud environments (AWS, GCP, Azure).
Demonstrated track record of leading technically ambiguous, cross-team infrastructure initiatives and driving them to measurable impact.
Strong architectural judgment and ability to make sound technical tradeoffs across performance, reliability, usability, and cost.
Willingness to travel to Sunnyvale, CA as needed.
Comfortable operating in highly ambiguous and dynamic environments.

Nice To Haves

7+ years of professional software engineering experience.
Deep expertise in PyTorch 2.x+ and distributed training frameworks.
Experience designing and developing training platforms that support FSDP, pipeline parallelism, and other scalable solutions for training large foundational models.
Experience profiling, analyzing, debugging, and optimizing training and data loading performance at scale.
Strong record of technical leadership through architecture reviews, roadmap influence, and cross-team execution.
Excellent communication skills, with the ability to build consensus, navigate controversial decisions, communicate risks clearly, and provide constructive technical feedback.
Self-motivated, execution-oriented, and motivated by delivering broad organizational impact.

Responsibilities

Define and drive the architecture, design, and development of scalable, reliable, and high-performance ML frameworks and platform capabilities to support model training at scale.
Lead model training performance analysis and optimization efforts across distributed training workflows, improving scalability, efficiency, and cost across heterogeneous hardware environments.
Raise the bar on system observability, debuggability, operational excellence, and developer experience across the ML training stack.
Own large, ambiguous, cross-functional technical initiatives from strategy through execution, including technical roadmap definition, tradeoff analysis, and delivery.
Influence platform direction by identifying long-term infrastructure investments, setting engineering standards, and driving adoption of best practices across teams.
Collaborate across organizational boundaries to align requirements, resolve technical disagreements, and integrate new capabilities into the platform ecosystem.
Mentor engineers through design reviews, technical guidance, and hands-on partnership, while elevating engineering quality across the team.