Machine Learning Engineer: Imitation and Reinforcement Learning for Robotics
Bedrock Robotics · San Francisco, CA
About this role
**Machine Learning Engineer: Imitation and Reinforcement Learning for Robotics** **About Bedrock** Bedrock is deploying autonomous systems on heavy construction equipment across the country. The team includes veterans from Waymo, Segment ($3.2B acquisition), and Uber Freight ($5B revenue). In two years, they've raised $350M and achieved the first fully autonomous excavator deployments in construction. **The Role** Develop and scale state-of-the-art learning architectures for behavior learning, with a focus on data-driven behavior policies and robust data infrastructure. **Key Responsibilities** • Design, train, validate, and launch models for behavior cloning and reinforcement learning • Build and maintain data ingestion, labeling, and management pipelines for high-quality training datasets • Build metrics to evaluate model performance in open loop, simulation, and real-world environments • Collaborate with simulation, systems, and infrastructure teams to integrate ML models into autonomous systems • Deploy and debug models in real-world environments, addressing latency, hardware constraints, and system integration **Required Experience** • 3+ years applying Machine Learning with Deep Learning frameworks (PyTorch/TensorFlow/JAX) • 3+ years building, deploying, and maintaining ML models in production • Familiarity with recent literature and methods in learned behavior policies • Practical experience in behavior cloning and/or reinforcement learning **Bonus Qualifications** • Experience with diffusion policies or Vision-Language-Action (VLA) models • Published work at ICRA, IROS, CoRL, CVPR, ECCV, ICCV, ICML, NeurIPS, or similar venues **Equal Opportunity Employer** Bedrock Robotics considers all qualified applicants without regard to protected characteristics and provides reasonable accommodations in the hiring process.
Listing freshness
CronJobs last confirmed this listing 1d ago. If its source stops confirming the opening for seven days, this page is removed from active inventory.