About The Position

You will design and ship the algorithms that decide how our trucks move. This spans classical motion planning and behavior, a growing learning-based prediction and planning pipeline, and the online mapping and map-building layer that planning depends on. You will work end to end — from problem framing and algorithm design through implementation, on-vehicle validation, and the cases that come back from the road. This is a hands-on engineering role for someone who wants their work driving real freight on public highways.

Requirements

  • B.S. or M.S. in Computer Science, Robotics, Electrical Engineering, Applied Math, or a related field, or equivalent practical experience.
  • 2+ years building production algorithms in robotics, autonomous vehicles, or a comparable real-time system.
  • Strong C++ and Python; comfortable owning performance-sensitive code that runs on-vehicle.
  • Solid foundation in at least one of: motion planning and optimization, prediction/behavior modeling, or mapping.
  • A track record of shipping algorithms that ran on real hardware or in production, not only in research or simulation.

Nice To Haves

  • Experience with learning-based components in a planning or prediction stack (sequence/trajectory models, imitation or reinforcement learning, model training and deployment pipelines, GPU inference in a real-time loop).
  • Experience with online/HD mapping, reference-line or lane-graph generation.
  • Familiarity with the planning–perception interface: uncertainty representation, occlusion reasoning, safety and liability evaluation, and agent modeling.
  • Experience operating in a safety-critical or heavily validated software environment.
  • Background in autonomous trucking or highway-speed autonomy.

Responsibilities

  • Develop and improve motion planning and behavior algorithms for highway and surface-street driving.
  • Build the learning-based pipeline for prediction and planning: data curation, model design and training, and integration on-vehicle.
  • Advance online mapping and map-building, and the real-time map signals that planning consumes.
  • Turn road and simulation cases into root-cause analysis, algorithmic fixes, and regression coverage that prevents recurrence.
  • Collaborate across perception, control, simulation, and operations to take features from design through validated release.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service