Senior Engineer, Machine Learning Application Developer

Samsung•San Jose, CA

About The Position

Samsung Austin Research and Development Center (SARC) and Advanced Computing Lab (ACL) are building a center of excellence for Intellectual Property (IP) applied to high-performance computing devices. As a Machine Learning Application Developer, you will develop neural rendering applications and machine learning (ML) software for efficient AI workload execution on Samsung's mobile GPUs. You will contribute to software solutions that bridge ML workloads and GPU hardware capabilities, optimizing performance, efficiency, and resource utilization for next-generation intelligent computing experiences. This role involves developing and optimizing neural rendering applications, API-level software, and ML operator implementations using Vulkan, OpenGL, and OpenCL. You will analyze software performance and hardware resource utilization to identify bottlenecks and optimize applications. Collaboration with hardware, software, and architecture teams is key to understanding hardware constraints and translating insights into optimized software. You will leverage low-level performance analysis techniques, including assembly-level investigation, to improve execution efficiency and maximize GPU utilization. You will take initiative on projects and advance best practices by staying current with advancements in ML, neural rendering, and GPU technologies.

Requirements

3+ years of experience with a Bachelor's Degree in Computer Science, Computer Engineering, or comparable field, or 2+ years of experience with a Master’s Degree, or Ph.D.
Strong programming skills in C, C++, and Python.
Proficiency with API-level programming using in Vulkan, OpenGL, OpenCL, and machine learning frameworks such as PyTorch and TensorFlow.
Understanding of GPU hardware architecture and experience with low-level performance profiling, analysis, and optimization.
Hands-on experience developing neural rendering applications at the API level.
Working knowledge of machine learning operators and workloads, including GEMM, convolution, activations, and related computational kernels.
Ability to analyze hardware resource constraints and bottlenecks and develop software optimizations that improve performance and efficiency.
Strong analytical and problem-solving skills, with the ability to identify bottlenecks and propose data-driven solutions.
Excellent communication and collaboration skills, with the ability to navigate ambiguity in a fast-paced, global team environment.

Nice To Haves

Working knowledge of assembly-level analysis, debugging, or optimization is preferred.

Responsibilities

Develop neural rendering applications and machine learning (ML) software for efficient execution of AI workloads on Samsung’s premium mobile GPUs.
Contribute to the development of software solutions that bridge machine learning workloads and GPU hardware capabilities.
Optimize performance, efficiency, and resource utilization to support next-generation intelligent computing experiences.
Develop and optimize neural rendering applications, API-level software, and ML operator implementations, including GEMM, convolution, activations, and related workloads, using Vulkan, OpenGL, and OpenCL.
Analyze software performance and hardware resource utilization to identify bottlenecks and optimize application performance, efficiency, and scalability across a variety of ML workloads.
Collaborate with GPU architects, software engineers, and hardware teams to understand underlying hardware constraints and translate performance insights into optimized software solutions.
Leverage low-level performance analysis techniques, including assembly-level investigation when needed, to improve execution efficiency and maximize GPU utilization.
Take initiatives on moderate-to-complex projects and help advance best practices and methodologies by staying current with the latest advancements in machine learning, neural rendering, and GPU technologies.