About the job
At Compound Eye, we empower machines to perceive their environment in three dimensions and in real-time, utilizing only passive sensors such as cameras and Inertial Measurement Units (IMUs). Our cutting-edge VIDAS™ technology transforms video from automotive-grade cameras into a rich, semantic 3D representation of the world, enabling vehicles and robots to navigate their surroundings—offering a fully redundant alternative to LiDAR and radar systems. We are a revenue-generating company delivering products to clients across the automotive, agriculture, healthcare, and defense sectors, backed by Khosla Ventures and other top-tier investors.
Our team of 20 professionals in the U. S. is on a mission to teach machines how to see. We cultivate a culture rooted in transparency, mutual respect, and the appreciation of innovative ideas, regardless of their origin.
The Role
We are in search of a Senior Computer Vision Engineer with extensive experience in state estimation, sensor fusion, and geometry-based perception. You will play a pivotal role at the heart of our VIDAS™ technology, developing and enhancing algorithms that enable machines to comprehend the world solely through camera data.
The most challenging problems we face may not have off-the-shelf solutions. You should have encountered scenarios where algorithms have failed, existing literature provided no guidance, and you had to revert to first principles to understand the underlying issues.
What You’ll Do
- Design and enhance algorithms for pose estimation, SLAM, Visual-Inertial Odometry (VIO), and 3D reconstruction.
- Develop and optimize navigation filters including Extended Kalman Filters (EKF) and other classical state estimation methodologies.
- Take ownership of sensor fusion pipelines that integrate IMU and camera data in Lidar-free configurations.
- Create and maintain multi-view geometry and camera calibration systems.
- Lead the development of real-time perception solutions for autonomous and robotic applications.
- Evaluate algorithm performance and failure modes, translating insights into actionable improvements.
- Guide and mentor a team of equally skilled engineers.
What You Bring
A minimum of 3–6 years of practical experience in computer vision or robotics perception, with a focus on:
- Pose estimation, SLAM, Visual-Inertial Odometry (VIO), and 3D reconstruction.
- Nonlinear optimization techniques (e.g., bundle adjustment, factor graphs).
- Classical state estimation methods and Kalman filters (EKF).
- Sensor fusion approaches integrating IMU and camera data.

