Leads the team you would join
Leads the work day to day
Owns the opening
From your own Gmail and LinkedIn.
About the role
This is a full-time onsite state estimation role in San Jose, California, for an engineer with at least four years of experience in multi-sensor fusion. You will own the architecture, algorithms, and calibration workflows for a data collection system that supports teleoperation and offline trajectory reconstruction for humanoid robots. The company builds general-purpose humanoid robots for home and commercial use. The work centers on real-time filtering, batch optimization, and fast calibration of human-worn sensors.
What you’ll do
- Design and implement real-time filters for streaming teleoperation and batch optimization for offline dataset generation, in modern C++.
- Develop and own subject-calibration routines that estimate body segment dimensions, joint offsets, and sensor-to-body extrinsics.
- Build sensor fusion architectures that combine spatial transforms, visual-inertial data, and inertial signals into full-body pose estimates.
- Address spatiotemporal calibration, environmental interference, and kinematic constraints on human skeletal models.
- Extract useful information from compliant tactile sensors that deform, stretch, or fold.
- Diagnose hardware limitations and inform future design requirements.
- Evaluate new sensing modalities for future hardware.
- Build diagnostic tools, validation pipelines, and error analysis workflows for online and offline algorithms.
What they’re looking for
- Four or more years building multi-sensor fusion and state estimation for dynamic hardware systems.
- Hands-on experience with (E)KFs, sliding-window estimators, factor graphs, GTSAM, Ceres, and non-linear least squares.
- Proven ability to design fast, reliable calibration, zeroing, and alignment workflows.
- Strong grounding in 3D kinematics, Lie groups (SE(3), SO(3)), forward and inverse kinematics, and constrained optimization.
- Ability to write high-performance, modular C++ for embedded or edge platforms, plus Python for analysis and visualization.
Nice to have
- Experience with low-latency streaming pipelines for teleoperation, haptics, or human-in-the-loop control.
- Background in human biomechanics, skeletal tracking, or body-mounted telemetry.
- Experience applying machine learning to motion priors, trajectory smoothing, or learned state estimation.
What’s on offer
- US base salary between $150,000 and $300,000 per year.
- Pay may vary based on job-related knowledge, skills, and experience, and total compensation may include additional components.
- Full-time, onsite in North San Jose, five days per week.
Skills
Questions about this role
What is the salary for this role?
The US base salary range is $150,000 to $300,000 per year, depending on factors such as skills and experience.
Is this role remote?
No, the posting requires five days per week in the San Jose office.
How many years of experience do I need?
The posting asks for four or more years of experience building multi-sensor fusion and state estimation solutions.