Description
You will be writing shaders that stress the GPU and cause high bandwidth traffic to and from the memory subsystem. You will apply your deep understanding of GPU architecture to write tests that stress the pipeline and exercise the interaction between the GPU and different IP blocks on the SoC. You will use the necessary driver and firmware APIs to compile shaders and prime operations to move data around in the SoC using the GPU threads/SIMD groups. You will understand the different levels of GPU memory hierarchy (Shared, Global, etc.) You will debug SoC software and hardware issues, and perform system level validation.
Minimum Qualifications
A minimum of a bachelor's degree in electrical engineering, computer engineering, or related field.
Preferred Qualifications
Experience with Graphics, OpenGL, and CUDA Experience with image processing, memory hierarchies, and compilers Strong C/C++ programming skills Implement and run ML/LLM workloads for GPU characterization Proficient in Python and GNU Make Deep understanding of compute systems, operating systems, and computer architecture Passionate about writing code close to the hardware, and debugging You enjoy the validation mentality: meticulous, curious, and persistent. Understanding of the Display pipeline is a plus. In-depth knowledge and experience in SoC system level validation and debugging of SoC software and hardware issues.
Learn more about this Employer on their Career Site
