Embedded Software Architect – Computer Vision

July 28, 2020
Santa Clara, California
Job Type


We are looking for an Embedded Software Architect – Computer Vision.

As the market leader in deep learning and parallel computing, NVIDIA is seeking an expert embedded software architect with strong experience in DSP and/or DSP SDKs to help us design, write and optimize computer vision software for NVIDIA's Programmable Vision Accelerator (PVA) engine in the Tegra SoC platform.

As a Software Architect, you will join a team of software engineers in creating highly optimized computer vision and machine learning kernels for specialized DSP hardware. You will use your design abilities, coding expertise, and creativity to help delivering state of the art real-time firmware and kernel mode drivers for a low power, high performance computer vision accelerator engine. You will work across organization to understand software requirements, efficiently map complex algorithms to the VLIW/SIMD hardware, integrate computer vision algorithms into a production firmware solution, and benchmark/optimize their performance.

You will be architecting and developing new features and improvements to realize the groundbreaking potential of NVIDIA mobile systems, ranging from self-driving cars, intelligent video analytics and autonomous mobile robotics. You will need to demonstrate an excellent technical leadership, communication, interpersonal, and analytical skills as well as a real passion for computer vision and a deep understanding of DSP software development. If this sounds like a fun challenge, we want to hear from you!

What you will be doing:

  • Create software architecture for future Nvidia’s Programmable Vision Accelerator (PVA) chips and enhance the functionality of currently shipping products.

  • Design and write custom embedded software for PVA engine to meet specific system and hardware requirements at the SoC level.
  • Help define forward-looking strategy and improvements to the PVA algorithms and system architecture.
  • Review hardware specifications and map algorithms to the architecture.
  • Ensure creation of extensive testing/validation for new hardware features in emulators and simulators.
  • Participate in the bring-up of the new generation of the world most advanced SoC.
  • Collaborate closely with other teams and software/hardware architects across Nvidia to support the architecture, design, creation, integration, and validation of PVA software under a common SoC umbrella.
  • Provide technical support and guidance for internal and external customers.
  • Mentor and guide technical development of the less experienced team members

What we need to see:

  • PhD or MS in Electrical Engineering, Computer Engineering, Computer Science, or equivalent experience

  • 10+ years of working experience in embedded industry, including 5+ years in technical leadership role
  • Deep understanding of SoC principles, general systems architectures, operating systems, device drivers, memory management, multithreading, and real-time scheduling.
  • Deep understanding and working experience with embedded architectures including DSP, computer vision, image/signal processing and robotics systems in general.
  • Excellent software development skills (C, C++) and outstanding problem-solving capabilities.
  • Proven expertise in architecting embedded software and development of highly optimized code for DSP, SIMD and/or VLIW processors
  • Experience with embedded Linux and/or QNX.
  • Outstanding interpersonal skills with ability to work in a global and diverse team operating in a fast-paced environment.
  • Good understanding of safety-critical software principles with experience in automotive or other highly regulated industries

Ways to stand out from the crowd:

  • Experience with ISO 26262 and IEC 61508 or equivalent quality/safety processes.

  • Understanding of software safety and safety development processes is a major plus.
  • Proven track record to effectively guide and influence in technically strong dynamic environment.
Drop files here browse files ...

Related Jobs