Machine Learning Engineer, Robotics
Job Description
Path Robotics is seeking a Machine Learning Engineer for its Robotics team in Columbus, OH (onsite). In this role, you will develop machine learning methods that improve how robots plan, scan, and act in real-world environments, including end-to-end deployment on live robotic cells.
Key Responsibilities
- Translate open-ended robotics problems across planning, scanning, and robot behavior into well-defined machine learning tasks.
- Train models that integrate with classical methods, including learned costs and initial guesses for trajectory optimization, plan quality and feasibility predictors, and learned collision models.
- Build viewpoint planning policies to select where and how a robot should scan, including prediction of scan quality prior to commitment.
- Apply imitation learning, reinforcement learning, and generative modeling approaches such as diffusion and flow matching when they outperform classical baselines.
- Create production-ready machine learning pipelines, including dataset creation from fleet logs and simulation, model training, and metric tracking.
- Evaluate models in the loop using offline metrics that estimate robot gains, perform failure-mode analysis, and implement calibrated confidence to trigger fallback to classical methods.
- Own model development end to end, moving from prototype to deployment on live robotic cells with low-latency inference and ongoing monitoring.
Required Qualifications
- Education: MS or PhD in Computer Science, Robotics, Machine Learning, or a related field.
- Experience: 2+ years of applied machine learning for robotics or autonomy, in research or industry.
- Proficiency in Python with PyTorch or JAX, including experience training, evaluating, and deploying deep learning models.
- Strong foundation in linear algebra, 3D geometry, kinematics, and optimization.
- Demonstrated experience running methods on real robot hardware.
- Ability to clearly explain model results and limitations so planning, perception, and field teams understand when to rely on the system.
Technologies
- Python, PyTorch, JAX
- C++, ROS 2
- CUDA, TensorRT, ONNX
- Isaac Lab, MuJoCo
Preferred Qualifications
- C++ experience and deployment of models into a C++ or ROS 2 production stack.
- Experience related to motion planning, trajectory optimization, or collision checking.
- Active perception and viewpoint strategies such as next-best-view or informative path planning, plus experience with 3D point cloud learning.
- Uncertainty estimation and calibration, or Bayesian and information-theoretic planning.
- GPU inference optimization using CUDA, TensorRT, or ONNX.
- Simulation and sim-to-real transfer experience using Isaac Lab and MuJoCo.
- Publications in venues such as RSS, CoRL, ICRA, IROS, NeurIPS, ICML, or similar.
Benefits
- Daily free lunch to stay fueled and connected with the team.
- Flexible PTO.
- Comprehensive medical, dental, and vision coverage.
- 6 weeks fully paid parental leave, plus an additional 6–8 weeks for birthing parents (12–14 weeks total).
- 401(k) retirement plan through Empower.
- Generous employee referral bonuses.