Senior Robotics Software Engineer, Motion Planning & Manipulation

About Valstad

Valstad is building the machine that builds the ships.

We combine industrial robotics, welding, sensing, and AI to turn 3D models into real high-mix ship structure. Our mission is to rapidly expand America's maritime output by 100x.

Our team works directly with large industrial robots and real manufacturing equipment at our Austin facility. The systems we build will ultimately be deployed across shipyards and industrial facilities throughout the country.

The Role

Valstad is seeking a Senior Robotics Software Engineer to build the motion-planning and manipulation systems that convert CAD geometry and manufacturing requirements into reliable robot execution.

You will work across the full planning stack, including coordinate frames, kinematics, inverse kinematics, collision checking, constrained Cartesian motion, trajectory generation, external-axis coordination, calibration, simulation, and industrial robot integration.

Our robots must execute long, process-constrained paths around large steel structures while maintaining precise tool position, orientation, speed, and clearance. They must account for linear rails, fixtures, workpiece variation, controller behavior, and imperfect real-world calibration.

This is not a simulation-only role or a position focused on configuring existing planning libraries. You will be expected to understand the underlying algorithms, build production-quality software, and debug the system on physical industrial robots.

What You Will Do

• Develop motion-planning systems for robotic welding, handling, scanning, inspection, and related manufacturing processes.

• Convert CAD-derived process paths into collision-free, executable robot trajectories.

• Generate constrained Cartesian paths while maintaining tool orientation, standoff, process angles, and travel speed.

• Develop strategies for inverse-kinematics selection and continuity along long manufacturing paths.

• Detect and avoid singularities, wrist flips, joint limits, collisions, and discontinuous robot configurations.

• Plan coordinated motion across industrial manipulators, linear rails, positioners, and other external axes.

• Develop reachability and placement-analysis tools for robots, workpieces, fixtures, and tooling.

• Incorporate controller limits, interpolation behavior, blending, acceleration, and process constraints into trajectory generation.

• Integrate robot, tool, fixture, workpiece, and sensor calibration into the planning system.

• Use measured workpiece geometry and sensor data to adjust nominal manufacturing paths.

• Develop planning and recovery strategies for operations that cannot be completed in a single robot configuration.

• Integrate generated trajectories with industrial robot controllers.

• Build simulation, visualization, log-replay, and regression-testing infrastructure.

• Diagnose discrepancies between planned, simulated, commanded, and physically executed motion.

• Work directly with automation, mechanical, welding, and manufacturing engineers during commissioning and production.

• Help define the architecture and technical standards for Valstad’s robotics platform.
  

What We Are Looking For

• 5+ years of professional experience developing robotics software, motion-planning systems, or manipulation algorithms.

• Strong modern C++ software-development skills.

• Deep understanding of rigid-body transformations, coordinate frames, and three-dimensional geometry.

• Strong knowledge of forward and inverse kinematics.

• Experience developing motion-planning or trajectory-generation systems for robotic manipulators.

• Experience with collision detection and constrained Cartesian motion.

• Experience debugging software on physical robotic systems.

• Ability to reason about numerical stability, geometric tolerances, and difficult edge cases.

• Strong software-engineering discipline, including testing, code review, debugging, and maintainable architecture.

• Ability to work effectively across robotics, controls, mechanical engineering, welding, and manufacturing.

• Willingness to work on-site and spend significant time with physical hardware.
  

Strongly Preferred

• Experience with industrial robots from FANUC, ABB, KUKA, Yaskawa, or similar manufacturers.

• Experience with ROS 2 and ROS-Industrial.

• Experience with MoveIt 2, Tesseract, OMPL, TrajOpt, Descartes, Drake, or similar frameworks.

• Experience implementing or modifying planning algorithms rather than only configuring existing packages.

• Experience planning for robots mounted on linear rails, gantries, or positioners.

• Experience optimizing redundant robotic systems with external axes.

• Experience maintaining continuous IK solutions along long process paths.

• Experience with trajectory optimization, sampling-based planning, or optimization-based inverse kinematics.

• Experience with SE(3), Lie groups, Jacobians, differential kinematics, or manipulability analysis.

• Experience with robot calibration, hand-eye calibration, TCP calibration, or workpiece localization.

• Experience integrating cameras, laser scanners, force-torque sensors, or other measurement systems.

• Experience with CAD or mesh geometry processing.

• Experience developing simulation, verification, or hardware-in-the-loop systems.

• Experience deploying robotic systems into production or customer environments.

• Experience with robotic welding, cutting, grinding, inspection, additive manufacturing, or another process-constrained application.

Shipbuilding experience is not required. We care more about deep robotics fundamentals, strong software engineering, and evidence that you can make complex robotic systems work reliably outside the laboratory.

What Success Looks Like

Valstad can reliably convert manufacturing intent and CAD geometry into safe, feasible, and executable robot trajectories.

Long manufacturing paths run without unexpected singularities, joint-limit failures, wrist flips, collisions, or discontinuous IK solutions. Calibration and sensor data flow cleanly into planning. Simulated behavior closely matches physical execution.

Planning failures are diagnosable rather than mysterious. Representative production scenarios are automatically tested. Engineers can reproduce hardware issues using logs and simulation.

Most importantly, Valstad’s robotic cells become less dependent on manual programming and individual expert knowledge. New structures can move from engineering data into production faster, with less intervention and greater reliability.