Machine Learning Engineer, Robotics

Path Robotics•Columbus, OH

About The Position

Path Robotics is seeking a Machine Learning Engineer to join its Robotics team and improve how robots plan, scan, and act. The role involves framing real-world robot problems as well-defined machine-learning problems, training models using data generated by production robotic cells, and deploying models into a production stack alongside classical planning and perception methods. Models run on live robots performing customer work. The engineer will own models from prototype through deployment on live robotic cells, building low-latency inference and monitoring. Work includes applying learning methods where they outperform classical baselines and evaluating model performance in the loop, with calibrated confidence and failure analysis to determine when systems should fall back to classical methods.

Requirements

  • MS or PhD in Computer Science, Robotics, Machine Learning, or a related field.
  • At least 2 years of applied machine learning for robotics or autonomy, in research or industry.
  • Fluency in Python with PyTorch or JAX, and experience training, evaluating, and deploying deep-learning models.
  • Strong foundations in linear algebra, 3D geometry, kinematics, and optimization.
  • Experience running methods on real robot hardware.
  • Ability to explain model results and limitations clearly to planning, perception, and field teams.

Nice To Haves

  • C++ and experience deploying models into a C++ or ROS 2 production stack.
  • Experience with motion planning, trajectory optimization, or collision checking.
  • Experience with active perception, next-best-view or informative path planning, or 3D point-cloud learning.
  • Experience with uncertainty estimation and calibration, Bayesian planning, or information-theoretic planning.
  • GPU inference optimization experience using CUDA, TensorRT, or ONNX.
  • Experience with simulation and sim-to-real transfer, such as Isaac Lab or MuJoCo.
  • Publications at RSS, CoRL, ICRA, IROS, NeurIPS, ICML, or similar venues.

Responsibilities

  • Frame open-ended problems in planning, scanning, and robot behavior as well-defined machine-learning problems.
  • Train models that work with classical methods, including initial guesses and learned costs for trajectory optimization, plan quality and feasibility predictors, and learned collision models.
  • Develop viewpoint-planning policies that choose where and how to scan and predict scan quality before the robot commits.
  • Apply imitation learning, reinforcement learning, and generative models such as diffusion and flow matching where they outperform classical baselines.
  • Build production machine-learning pipelines, including datasets from fleet logs and simulation, training, and metrics.
  • Evaluate models in the loop using offline metrics that predict gains on the robot, failure-mode analysis, and calibrated confidence to determine when systems fall back to classical methods.
  • Own models end to end, from prototype to deployment on live robotic cells, with low-latency inference and monitoring.

Benefits

  • Daily free lunch.
  • Flexible PTO.
  • Comprehensive medical, dental, and vision coverage.
  • Six weeks of fully paid parental leave, plus an additional 6–8 weeks for birthing parents (12–14 weeks total).
  • 401(k) retirement plan through Empower.
  • Employee referral bonuses.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service