Staff ML Engineer - Embodied AI Scaling Foundations

Responsibilities: Design experiments linking data mixtures and curricula to model behavior Apply self-supervised pre-training, imitation learning, reinforcement learning, and fine-tuning to driving tasks Develop data curation and mining methods (auto-labeling, deduplication, long-tail discovery) to improve training value Define offline metrics that predict on-road behavior and guide data/model decisions Trace model failures to data root causes and specify needed data fixes Train models at scale on multi-GPU/multi-node pipelines with platform team collaboration Collaborate with cross-functional teams to deploy models in onboard systems and share learnings

Key requirements: Master's or PhD in a relevant field Strong ML fundamentals with rigorous experimental design Proficiency in Python and PyTorch, with experience on large datasets Hands-on data-centric ML experience (curation, sampling, labeling, evaluation) Knowledge of large-scale foundation models and pre-training/fine-tuning Data analysis skills (NumPy, Pandas; SQL or Spark) Ability to deliver ML results under real-world constraints and communicate clearly Clear communication to both engineers and non-experts Collaborative mindset for cross-functional work Result-oriented and deadline-driven approach Python PyTorch NumPy