Staff MLOps Engineer
Core
Design and own the MLOps platform layer for Apptronik's Apollo humanoid robot, managing the full lifecycle from dataset versioning and experiment tracking to model registry, evaluation, and deployment to robots in the field.
Role type
Staff MLOps Engineer (Hands-on IC with technical leadership)
Builds
The MLOps platform serving the Apollo robot fleet, connecting teleoperation data to deployed autonomy.
Domain
Embodied AI / Humanoid Robotics / Robotics Infrastructure
Deliverable
production ML models
Required skills
Python, Go/Rust/C++, MLOps platform ownership, Kubernetes, cloud infrastructure (AWS/GCP/Azure), Docker, Git, CI/CD, dataset versioning (DVC/LakeFS/Delta), experiment tracking (MLflow/W&B/Determined), model registry, policy serving, evaluation framework design, technical architecture, cross-team influence, mentorship
Preferred skills
Edge/embedded ML deployment (ONNX/TensorRT), RL training infrastructure, humanoid robotics/teleoperation domains, simulation-in-the-loop evaluation (IsaacSim/MuJoCo), policy gating/shadow deployment, open-source MLOps contributions
Technologies
Kubernetes, AWS, GCP, Azure, Docker, Git, MLflow, W&B, Determined, DVC, LakeFS, Delta, ONNX, TensorRT, torch.compile, IsaacSim, MuJoCo
Responsibilities
Define subsystem interfaces and engineering standards for the MLOps platform; serve as the primary technical point of contact for Autonomy, Data Platform, and TeleOp; design and operate the dataset layer end-to-end including versioning and lineage; build and operate a first-class model registry with approval workflows; define automated evaluation benchmarks and metrics frameworks; own the path from registered model to on-robot inference including packaging and observability; mentor mid-level and senior engineers on the MLOps team.
Seniority
Staff, hands-on IC with technical leadership