Staff Software Engineer, ML
Core
Design and own the architecture of the MLOps platform connecting teleoperation data to deployed autonomy on Apollo robots, managing dataset lifecycle, model registry, evaluation harnesses, and on-robot serving paths.
Role type
Staff MLOps Engineer (hands-on IC)
Builds
MLOps platform layer (dataset versioning, experiment tracking, model registry, evaluation, and robot deployment pipelines)
Domain
Robotics + Applied AI
Deliverable
production ML models
Required skills
Python, Go/Rust/C++, MLOps platform ownership, Kubernetes, cloud infrastructure (AWS/GCP/Azure), CI/CD, model lifecycle management, technical leadership
Preferred skills
Edge/embedded deployment (ONNX/TensorRT), RL training infrastructure, humanoid robotics, simulation-in-the-loop evaluation, policy gating strategies
Technologies
Python, Go, Rust, C++, Kubernetes, Docker, Git, MLflow, W&B, DVC, LakeFS, Delta, ONNX, TensorRT, IsaacSim, MuJoCo
Responsibilities
Define subsystem interfaces and architecture decisions for the MLOps platform; Design and operate dataset versioning and lineage systems; Build and manage model registry with approval workflows; Develop automated evaluation and qualification frameworks; Own the path from registered model to on-robot inference; Mentor mid-level and senior engineers; Partner with Training Infrastructure on cluster contracts
Seniority
Staff, hands-on IC
