Computer Vision Engineer

RetailNext is the worldwide market leader and expert in brick-and-mortar retail analytics. We track more than a billion shopping trips per year by collecting data from nearly 100,000 sensors in retail stores and analyze trillions of data points annually. Headquartered in San Jose, CA, RetailNext is a growing global brand operating in more than 70 countries. Cutting-edge Computer Vision is at the core of the RetailNext’s suite of proprietary technologies.

Our R&D team works on problems ranging from image registration, mosaicking, and 3D reconstruction to multiple object tracking and facial analysis. We recently branched out from our roots as a software company by designing and building our own cloud-managed stereo sensor, Aurora, that features onboard shopper detection and tracking analytics using video, in addition to WiFi and Bluetooth-based analytics and on-board HD video recording.

The need to understand shopper behavior drives us to research and develop theoretically and computationally/practically efficient algorithms for video analytics, and we’re looking for engineers excited by such problems. This engineer will have a strong understanding of multi-view geometry, an inventive eye for developing new techniques for our unique challenges and a desire to efficiently scale computer vision into large installations around the world.

The position requires a Master’s Degree in Computer Science, Computer Vision, or Machine Learning. Strong programming skills in C/C++ and rapid prototyping experience in Python or MATLAB is required.

Responsibilities

Design, implement, validate, and release applications and capabilities in C++ and Python
Develop robust software for integrating multiple sensors and tracking systems
The candidate will lead the development of new algorithms, and work on improving the performance and efficiency of current algorithms

Qualifications

Master’s Degree in Computer Science or a closely related field, or equivalent experience
Experience with Computer Vision, Multi-view Geometry, 3D Vision, SLAM, or Structure from motion
C/C++ experience, and excellent software engineering skills
Experience and/or education with linear algebra
Experience hands-on implementation and product development experience for real-time sensing problems
Experience developing for performance at both the algorithm and code levels
Knowledge of multi-view camera systems, 3D reconstruction, bundle adjustment, intrinsic and extrinsic calibration methods, SfM using low-level image features, e.g., SIFT