Our client specializes in enabling machines to perceive their environment in 3D and in real time by utilizing only passive sensors, such as cameras and inertial measurement units (IMUs). Their innovative technology processes video from automotive-grade cameras to generate a detailed, semantic 3D representation of the world, which vehicles and robots use for navigation. The company works on a wide variety of platforms including cars, trucks, drones, aircraft, and more and serves various sectors, including automotive, agriculture, healthcare, and defense.

While their headquarters are in Silicon Valley, this position is fully and permanently remote.

RESPONSIBILITIES:

As a Computer Vision Engineer, you will play a key role in the design and development of advanced technologies and products such as real-time perception, SLAM, 3D reconstruction, and visual-inertial odometry (VIO). Your responsibilities will include:

Contributing to the development of cutting-edge techniques and algorithms for robotic perception.
Writing production-level C++ code for various platforms, including GPUs.

QUALIFICATIONS:

Proficient in developing and fine-tuning math-intensive computer vision algorithms—such as calibration, pose estimation, and sensor fusion (rather than simply relying on existing libraries).
A solid foundation in linear algebra and nonlinear optimization
Experience with classical computer vision techniques (understanding the math behind the CV algorithms, able to implement them from scratch, and analyze the performance if needed).
Familiarity with real-time perception, 3D reconstruction, visual-inertial odometry, and SLAM are good to have
Extensive experience in writing high-quality, production-level C++ code
Familiarity with parallel programming frameworks like CUDA, Metal, or OpenCL
B.S. or higher in Computer Science, Computer Engineering, or related fields
U.S. citizen or permanent resident (green card holder)