Machine Learning Engineer - Robot Learning (Reinforcement Learning)

Build the Path Forward

At Path Robotics, we're building the future of embodied intelligence. Our AI-driven systems enable robots to adapt, learn, and perform in the real world, closing the skilled labor gap and transforming industries. We go beyond traditional methods, combining perception, reasoning, and control to deliver field-ready AI that is risk-aware, reliable, and continuously improving through real-world use.

Big, hard problems are our everyday work, and our team of intelligent, humble, and driven people make the impossible possible together.

Path built its business on autonomous welding, with robots deployed at customer sites across North America. Now our Robot Learning team is building what comes next: a general-purpose robotic foundation model for manufacturing, trained at scale with reinforcement learning and deployed on real factory floors.

Why This Role

Build a foundation model for manufacturing: Manufacturing demands far more: heavy parts, precision measured in millimeters, and steps that can't be undone once they're done. No existing model meets that bar, so we're building one. You'll help create a model that learns from real factory data and can take on new tasks and new parts without being reprogrammed each time.

Scale RL: Imitation learning gets robots to human level. Our customers need more: faster cycle times, higher throughput, and precision humans can't sustain. We're scaling RL with massively parallel simulation, transferring policies to real industrial robots, and building toward one generalist model that learns from many specialist policies. You'll work on open problems in RL and see the results run in production at customer sites.

What You'll Do

• Train RL policies for high-precision manipulation in large-scale, GPU-parallel simulation.
• Transfer policies from simulation to real industrial robots, using vision and force/torque sensing.
• Scale from task-specific policies to generalist models that handle new parts without retraining.
• Combine foundation models (VLAs, world models) with RL to solve long-horizon assembly tasks.
• Deploy and evaluate policies on real robotic cells with customers, and use the results to improve the models.

Who You Are

• MS or PhD in Robotics, Computer Science, or a related field.
• Hands-on experience applying RL to manipulation or contact-rich control.
• Experience with sim-to-real transfer and GPU-parallel simulation (e.g., Isaac Lab, MuJoCo).
• Experience deploying learned policies on physical robots.
• Strong Python and PyTorch or JAX skills, including training at scale.