Leads the team this role joins
In-house, owns the opening
Put the opening up
From your own Gmail and LinkedIn.
About the role
This is a full-time onsite 3D vision role in San Francisco for an engineer who builds the perception systems that help robots understand and move through real environments. The role combines classical geometry with deep learning to take research ideas into production-quality systems. The Bot Company is building a general-purpose home robot and runs a small, lean engineering team. Candidates need strong Python or C++ and hands-on experience building 3D vision systems from scratch.
What you’ll do
- Design and build end-to-end 3D perception systems for real robotic environments.
- Develop geometric vision pipelines, including SLAM, reconstruction and tracking, and integrate them with learned models.
- Implement and optimize deep learning models for depth, optical flow, correspondence and 3D understanding.
- Prototype quickly, measure performance in realistic conditions, and iterate toward production quality.
- Work with robotics, simulation and platform teams to integrate perception across the full stack.
What they’re looking for
- Very strong coding in Python and or C++.
- Experience with PyTorch and CUDA, including custom operators, optimized data pipelines or model acceleration.
- Experience building 3D vision systems from scratch rather than only using off-the-shelf libraries.
- Solid understanding of camera, IMU, wheel odometry and lidar measurement models, calibration, time synchronization and noise.
- Deep expertise in at least two of: multi-view geometry, bundle adjustment, pose graph optimization and visual-inertial odometry; dense 3D reconstruction and novel-view synthesis; training depth, optical flow or feature matching networks.
- Ability to move quickly and independently on hard geometry problems.
What’s on offer
- Full-time, onsite in San Francisco.
- Scope and direct ownership within a small team.
Skills
Questions about this role
Is this role remote?
No. The role is onsite in San Francisco.
Which skills are required?
Strong Python or C++, PyTorch and CUDA, and deep expertise in at least two areas of 3D vision such as multi-view geometry, dense reconstruction or depth and flow networks.