Apple is looking to engineer continuous improvements in stability and performance for private cloud compute, as well as implement new functionality for generative AI as part of the Private Cloud Compute component of Apple Intelligence.
Requirements
- Experience programming in: Swift, C, C++, iOS/macOS or XCode
- Practical experience running machine learning models and evaluating them for quality and performance metrics
- familiar with Apple ML stack (ANE, CoreML, MPS/Metal)
- high-level general distributed ML stack (PyTorch-distributed, NCCL)
- high throughput inter-chip communication systems
- on-device iOS development
Responsibilities
- engineer continuous improvements in stability and performance for private cloud compute
- implement entirely new functionality as it emerges from the research community
- write performant and scalable frameworks (in Swift and C++) to distribute and coordinate ML inference tasks to different hardware acceleration IP blocks on different SoCs
- integrate inference code into a full service stack to ensure that user traffic is served reliably and performantly
- developing code that is easy and safe to develop, update and monitor
Other
- Experience working as a software engineer on large production systems
- produce reliable, maintainable, deliverable software
- Comfortable diving deep - working across multiple levels of abstraction
- Good at handling relationships & communication
- collaborate well with colleagues across a wide range of functions
