Engineer, GPU Software Compute

Samsung•San Jose, CA

1d•$116,000 - $174,000•Onsite

About The Position

Samsung, a world leader in advanced semiconductor technology, is founded on a simple philosophy – the endless pursuit of excellence will create a better world for all. At Samsung Austin Research and Development Center (SARC) and Advanced Computing Lab (ACL), we are building a center of excellence for Intellectual Property (IP) that is applied to high-performance computing devices (mobile, automotive, and other custom market segments) consumed by millions of people around the world. Come build with us! Role and Responsibilities As a Software Driver Engineer, you will join our GPU Software team to design and implement user-mode driver features that enable high-performance compute workloads and application development for Samsung’s premium mobile GPUs. Your work will directly impact the GPU software development pipeline and empower developers to unlock the full potential of GPU compute and enable exceptional performance and quality across leading-edge mobile applications. As a successful candidate, you will have hands-on experience with parallel programming (OpenCL or CUDA or Vulkan) and knowledge of GPU or other parallel processing architectures. You are an innovator. You are skilled at developing tools used for driver and application development. You take ownership in delivering high quality results—from debugging and troubleshooting driver and application issues, to maintaining the Compute driver build and execution environments. You’re curious and forward-thinking . You challenge conventional methods with data-driven insights and innovative approach and always seek for better ways to solve complex problems and improve performance. You’re a team player . You thrive in cross-functional collaboration and partnering with software and hardware teams to achieve optimal GPU performance. You are a lifelong learner. You enjoy staying up-to-date with the latest technologies, continuously sharpening your competencies and knowledge through learning and experimentation. Our Team As the first to introduce Ray Tracing in the mobile platform, we collaborate with game engine teams to seamlessly integrate console-quality features to provide users a flawless mobile gaming experience. We pride ourselves on our ability to work cohesively and efficiently, continuously pushing the limits of technology with our next-gen GPU software/hardware codesign. With a focus on post-silicon performance optimization and workload analysis, we ensure our graphic software is the best in class. At SARC/ACL, you will work alongside a diverse global task force with cross-functional impact, contributing to multiple stages of the technology development cycle and seeing your work in real-world products. Here your bold ideas are encouraged, learning always happens, and teamwork powers our innovations.

Requirements

Bachelor's Degree in Computer Science, Computer Engineering, or comparable field
Strong programming skills in C, C++, and Python, with experience implementing 3D graphics features, GPU hardware, and firmware development.
Good understanding of GPU hardware, ray tracing, rasterization, and linear algebra, as well as experience with low-level performance profiling and optimization.
Hands-on experience in debugging complex issues in multi-threaded environments and navigating operating system fundamentals and concepts.
Familiarity with parallel programming APIs and languages, such as OpenCL or CUDA or Vulkan, and have experience with game engines (e.g., Unreal/Unity).
Background in Linux and Android development, with knowledge of offline and JIT compiler designs.
Availability to work on-site 5 days a week
ability to access information subject to U.S. export control restrictions

Responsibilities

design and implement user-mode driver features that enable high-performance compute workloads and application development for Samsung’s premium mobile GPUs
develop tools used for driver and application development
debugging and troubleshooting driver and application issues
maintaining the Compute driver build and execution environments
cross-functional collaboration and partnering with software and hardware teams to achieve optimal GPU performance