SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Dyna Robotics is seeking a Fleet Reliability Engineer to own the health and performance of its deployed humanoid robot fleet. This is a hands-on role spanning hardware diagnostics, firmware troubleshooting, mechanical design, and data infrastructure.
You will root-cause complex failures across the fleet—actuator dropouts, encoder drift, thermal events, CAN bus faults, and control-stack bugs—distinguishing genuine hardware degradation from firmware, calibration, and model-side issues. The role demands comfort at the bench: soldering, reworking electronics, swapping components, and running physical validation with test fixtures you design and build.
Key responsibilities include designing and fabricating mechanical test fixtures, jigs, and brackets; translating mechanical evidence (backlash, wear, preload loss) into concrete design changes; building fleet health metrics and alert rules tuned to real faults; developing calibration and test tooling with safety gates; extracting and decoding robot telemetry at scale (MCAP, npz, episode archives); and writing runbooks and SOPs for handoff to teammates.
You will work at the intersection of hardware, firmware, controls, and data infrastructure, collaborating closely with mechanical and controls teams. You'll review PRs across firmware defaults, control code, and deployment configuration, catching interaction bugs between calibration tools, boot checks, and safety gates.
Required: 3–5 years in robotics, mechatronics, hardware test, or reliability engineering with significant fielded-hardware experience. You must be genuinely hands-on with soldering iron, multimeter, and calipers; understand actuator failure modes; hold your own in design reviews; and have working knowledge of robot control (impedance, torque control, controller state lifecycles). Strong Python for data analysis, solid time-series and signal-statistics instincts, fluency with actuators and motor control (quasi-direct-drive, encoders, CAN), and experience with observability stacks (Datadog or similar) are essential. Rigor about evidence and a bias toward small, verifiable scope are non-negotiable.
Bonus: teleoperation/VR experience, hardware qualification (EVT/DVT/PVT) background, MuJoCo simulation, or a track record of building tools and SOPs that teams adopt.
