Computer Vision Architect
Sunnyvale, California, United States
Machine Learning and AI
The Video Computer Vision org is a centralized applied research and engineering organization responsible for developing real-time on-device Computer Vision and Machine Perception technologies across Apple products. We balance research and product to deliver Apple quality, state-of-the-art experiences, innovating through the full stack, and partnering with HW, SW and ML teams to influence the sensor and silicon roadmap that brings our vision to life.
We're looking for highly self-motivated and hard-working individual with expertise in machine learning and computer vision to join our team to develop new technologies for Apple’s next generation products. Are your ready to be apart of the next big thing at Apple?
Key Qualifications
- Ability to evaluate existing and emerging imaging systems against use case requirements, including, understanding of hardware engines (CPU, GPU, Neural Network Accelerator, DSP, etc.)
- Understanding of difference between classes of algorithms and how to optimize their efficiency on available hardware engines
- Understanding of the impact processing flow selection will have on overall system performance, power consumption and resource sharing by various algorithms.
- Deep knowledge of sensors and camera technologies, including, image acquisition (sensor+lensing+ISP), rolling vs global, color vs monochrome, rectilinear vs equidistant and their impacts on algorithms, color management, performance optimization in low light conditions, etc.
- Understanding of system resources supporting image collection and camera calibration in multi-camera systems and ability to define common system pre-processing blocks for camera streams.
- Showed strength in system architecture and performance analysis, including, input interface definition/optimization of image streams serving multiple algorithms, estimation of system compute requirements based on required performance.
- Ability to define system-wide policies optimizing resources in multi-sensor streaming systems with memory constraints
- Ability to translate product goals to feature level architectures and the ability to derive for each feature the end-to-end requirements (with rationale) and decompose them down to module/component specific requirements.
- Ability to work multi-functionally (Algorithms, Embedded, ML, Mechanical, Electrical, Optical, Human Factors, Vision science etc.)
Description
The ideal candidate must have proven track record of having participated in the development of multiple consumer electronic products. Should have broad understanding of hardware and software technologies that are employed in low-power hand-held / wearable devices. Must be a self-starter who is highly-motivated and an organized thinker with excellent communication and presentation skills.
Education & Experience
BS and a minimum of 10 years relevant industry experience