3D Vision Researcher / Engineer

About the role

We are looking for a 3D Vision Researcher / Engineer to design and implement state-of-the-art perception systems that enable our home robots to understand and navigate real-world environments. You will work in a small, high-impact team in San Francisco, turning cutting-edge research into production-ready pipelines.

Key responsibilities

• Design and build end-to-end 3D perception pipelines, including SLAM, reconstruction, and tracking.
• Develop and integrate geometric vision algorithms with deep-learning models for depth, optical flow, and feature correspondence.
• Prototype rapidly, evaluate performance in realistic robotic settings, and iterate toward production quality.
• Collaborate with robotics, simulation, and platform teams to embed perception into the full robot stack.
• Maintain and optimize codebases in Python and C++ for speed and scalability.

Required profile

• Exceptional mental acuity and ability to learn and reason quickly across unfamiliar domains.
• Strong engineering curiosity and a track record of building complex systems from scratch.
• High-performance mindset: comfortable with ambiguity, fast-paced environments, and delivering results.

Required skills

• Proficient programming in Python and C++.
• Experience with PyTorch and CUDA, including custom operators and model acceleration.
• Deep knowledge of multi-view geometry, bundle adjustment, pose-graph optimization, and visual-inertial odometry.
• Expertise in dense 3D reconstruction, novel-view synthesis (e.g., NeRF, Gaussian Splatting).
• Hands-on experience with sensor modalities such as cameras, LiDAR, IMU, and wheel odometry, including calibration and synchronization.

What we offer

• Opportunity to work on cutting-edge robotics perception in a lean, fast-moving team.
• Direct impact on products that will reach millions of homes.
• Competitive compensation and equity in a growing startup.