Watney
Staff Machine Learning Engineer
San Francisco · On-site · Staff · Full-time · Posted Sep 16, 2026
About the job
Build and deploy machine-learning models for autonomous robots, using fleet data to improve real-world performance across perception, planning, navigation, and manipulation.
What you'll do
- Own model quality end-to-end, including dataset curation, training, and evaluation.
- Train and evaluate policies using multimodal sensor data and experiments with data, features, and architectures.
- Deploy models to production and build low-latency, 24/7 inference systems for the robot fleet.
What they ask for
- Experience training large-scale multimodal models on GPU fleets, using techniques such as video semantic segmentation, 6D pose estimation, VLAs, and BEVs.
- Experience owning datasets end-to-end, including decisions about data collection, labeling, and selection.
- Production training and inference coding experience with Python and at least one of PyTorch or JAX, plus Triton.
- Experience with cluster management systems, including Ray, Slurm, or Kubernetes.
Tools
Cameras, IMUs, LiDar, Odometry, Python, PyTorch, JAX, Triton
AI focus
Train, evaluate, and deploy multimodal models and policies for autonomous robotics, improving real-world fleet performance.
Model-written summary · the original posting is authoritative