Staff Software Engineer, ML Training Infrastructure

Role overview Set technical direction and build core infrastructure for a multi-tenant machine learning training platform. The work spans the AI development workflow, from data preparation and model configuration through training, validation, profiling, and acceleration, helping machine learning teams iterate more quickly and reliably on systems used in autonomous vehicles.

Responsibilities - Design and evolve platform components for orchestration, training abstractions, control-plane operations, observability, and performance tuning. - Build end-to-end model pipelines covering log processing, feature extraction, dataset schemas and storage, model configuration, training, and profiling. - Investigate infrastructure performance and remove bottlenecks affecting training and iteration speed. - Provide reusable abstractions and tools that help machine learning engineers work independently and efficiently. - Integrate open-source tools for profiling and optimization, and encourage effective engineering practices across teams. - Partner across platform, infrastructure, autonomy, and safety evaluation groups while balancing hands-on engineering with longer-term technical direction.

Requirements - Bachelor’s or master’s degree in computer science, engineering, or a related field. - At least six years of experience with machine learning platforms and machine learning applications. - Strong programming skills in Python, C++, or an equivalent language. - Experience with a machine learning or data platform such as Lance, PyTorch, or Ray Data, or comparable tools. - Track record building scalable, reliable infrastructure with engineers across multiple modeling teams. - Experience with model training, model optimization, or large-scale data processing, plus the ability to explain technical tradeoffs and build cross-functional alignment. - Strong analytical, problem-solving, written, and verbal communication skills.

Nice to have - Autonomous vehicle experience; modeling experience is also a plus.