Senior AI/ML Performance Engineer

About the role

Join a capacity and performance engineering group that shapes strategy for large scale machine learning infrastructure and keeps training and inference environments efficient and reliable.

What you will do

• Run autonomous vehicle models to guide long term GPU system strategy and infrastructure roadmaps
• Analyze production workloads in depth to find bottlenecks and propose high impact optimizations
• Work with research, infrastructure and cloud vendor teams on projects that raise velocity and cut cost
• Spot architectural improvements that keep large scale training and inference scalable and dependable

What they look for

• At least five years of professional work in high scale infrastructure or machine learning systems
• A computer science degree or a related technical field, or comparable practical experience
• Expert Python coding and the ability to design and debug within the PyTorch ecosystem
• A record of fixing performance problems in large distributed production settings
• Solid grasp of distributed systems, modern machine learning design and high performance computing
• Hands on Kubernetes experience plus familiarity with Nvidia DCGM, nvidia-smi and Grafana

Nice to have

• Eight or more years of relevant industry experience
• Knowledge of enterprise Nvidia GPU architectures such as H100, B200 and GB200
• Experience scaling open source models through the Hugging Face ecosystem
• BigQuery skills for large scale analysis and reporting