Staff Robot Learning Engineer, VLA & Bimanual Manipulation
Skills, Inc.
Mountain View, CA, United States
3 days ago
- Discuss this with your agent
- Open in Claude
- Open in ChatGPT
Apply on jobs.localjobnetwork.com
Prepare application
- Draft this with your agent
- Open in Claude
- Open in ChatGPT
Role details
Contract type
Temporary contract
Employment type
Full-time (> 32 hours)
Experience level
Experienced
Experience required
3 years minimum
Compensation
$291,200
Working hours
Regular working hours
Job source
Tech stack
Big Data
Python (Programming Language)
Kinematics
Machine Learning
Motion Planning
Reinforcement Learning
Pytorch
Information Technology
Free and Open-Source Software
Job description
Join the robotics and AI innovation lab of a major automotive manufacturer as a senior, hands-on robot learning engineer. You’ll adapt, fine-tune and deploy Vision-Language-Action (VLA) models and world models on physical robots for industrial manipulation: kitting, bin-picking, assembly, and coordinated dual-arm tasks. This is an individual contributor role with real technical influence. You’ll shape model and architecture decisions, guide other engineers, and report results directly to engineering and executive leadership.
Mountain View, CA (on-site, 5 days a week)
What you’ll do
- Evaluate, fine-tune and deploy VLA models (Pi0/Pi0.5, OpenVLA, GR00T or similar) on physical robots
- Build learning-based bimanual manipulation: object handoffs, contact-rich tasks, coordinated dual-arm motion, multi-step tasks
- Apply world models, World Action Models and learned dynamics to improve planning and policy adaptability
- Integrate learned models with perception, motion planning, robot control and inference infrastructure
- Run real-robot experiments, benchmark rigorously, analyze failures, and drive gains in success rate, cycle time, robustness and generalization
- Shape data collection, teleoperation, dataset quality standards and model post-training
- Work closely with reinforcement learning, simulation and controls teams, and provide technical guidance to engineers including an RL specialist
- Report progress, results and risks to engineering and executive leadership in clear, data-driven updates
Requirements
- PhD in Robotics, Machine Learning, Computer Science, EE, ME or a related field, plus 3+ years of post-PhD research or industry experience in robot learning
- At least one first-author paper at a top venue (CoRL, RSS, ICRA, IROS, NeurIPS, ICML, ICLR, CVPR or similar). Real-world deployment experience counts for more than publication count.
- Hands-on experience fine-tuning, evaluating and deploying VLA models on physical robots
- Experience developing or deploying bimanual manipulation on physical robots
- Experience with at least one of: world models, World Action Models, learned dynamics or model-based robot learning
- Strong grasp of robotic manipulation, kinematics, coordinate frames, motion planning and control
- Python and PyTorch, plus working knowledge of ROS/ROS 2
- Ability to turn technical results into a clear story for business and manufacturing leaders
Nice to have
- Reinforcement learning, imitation learning, residual policies or policy adaptation
- Teleoperation, large-scale data collection and multimodal datasets
- Industrial or collaborative robots (Fanuc, Universal Robots)
- Open-source contributions or robot learning systems running in production
Apply for this position
This job is hosted externally. Click below to view the full posting and apply.
Apply on jobs.localjobnetwork.com
Prepare application
- Draft this with your agent
- Open in Claude
- Open in ChatGPT
Good distractions
Loading talks and stories from around this role…