Apple Silicon GPU Driver Engineer - Performance at Apple (Cupertino, CA)

Apple Silicon GPU Driver Engineer - Performance at Apple (Cupertino, CA)


Add To Bookmarks
Company:
Location: Cupertino, CA
Type: Full Time
Created: 2021-09-04 05:01:04

Apply Here


Apples GPU Software team develops Metal, an API for GPU programming that allows efficient use of the GPU and minimizes CPU overhead. It has enabled developers to achieve far greater realism, detail, and interactivity in their games. It is the foundation for utilizing the GPU across all of Apple's innovative products, from iPhone, iPad, and iPod touch, to Apple TV, and macOS, as well as Apples professional and consumer applications from Final Cut and Aperture to iLife and iWork. We are looking for an engineer to join the Apple Silicon GPU Driver Performance Team, which is responsible for helping to make sure Apple Silicon GPUs run blazingly fast. The ideal candidate will have a passion for squeezing the best possible performance out of the GPU, and wont be afraid to get down into the low level details of the GPU architecture in order to do it.The Apple Silicon GPU Driver Performance Team is responsible for ensuring great GPU performance across all of our products. The teams responsibilities include:* Defining and developing the driver software architecture for GPU introspection capabilities (including performance counters, shader profiling, tracing, etc) to enable upper level software and tools to measure and tune GPU workloads. This includes working at the kernel and firmware level, and up the stack to the user API level.* Working with 1st and 3rd party developers on tuning GPU workloads, algorithms, and shaders to achieve the best possible performance on Apple - platforms.* Helping to drive and refine both software and hardware architecture to maximize GPU performance and software introspection capabilities.* Working with internal hardware teams to ensure that our hardware roadmap continues to deliver best in class mobile GPU performance for both graphics and compute workloads.