ML Platform Architect – Apple Silicon Performance

Apple, Inc.; Cupertino (CA); In loco; USD 185.000 - 278.000; Tempo pieno

Vantaggi offerti da questo lavoro

Medical and dental Coverage

Retirement benefits

Employee stock programs

Relocation assistance

Tuition reimbursement

Discounted products

Descrizione del lavoro

Apple’s Platform Architecture group in Cupertino is seeking an engineer to map ML workloads to Apple silicon. This onsite role collaborates with cross-team engineers to build optimized ML workload implementations and to develop performance models and simulations that guide architecture decisions.

You will communicate findings to software and algorithm teams to validate trade-offs and improve execution efficiency, and you will prototype algorithms on CPU, GPU, and Neural Engine.

Competenze

• Bachelor’s degree in a relevant field
• Programming experience in C/C++ and/or Python
• Experience in computer architecture
• Experience with at least one hardware IP, such as ML hardware accelerators or processing units including GPUs, image/video, CPUs, or similar

Mansioni

• Create optimized implementations of ML workloads on Neural Engine, GPU, and CPU running on Apple silicon
• Partner with IP and SoC architecture teams to develop performance models and simulations for future hardware
• Run performance studies to inform and validate architecture decisions
• Work with system teams to build high-level performance models for emerging ML techniques and evaluate system architecture trade-offs
• Explore multiple approaches to mapping ML workloads onto Apple silicon and produce supporting performance models and simulations
• Share results and learnings with software and algorithm teams

Conoscenze

C/C++

Python

Verbal and written communication

Performance modeling

ML frameworks

Formazione

Bachelor’s degree

MS or PhD in EE/CE/CS or related field

Strumenti

PyTorch

Compiler frameworks/technologies

Neural Engine

GPU

CPU

Descrizione del lavoro

Apple’s Platform Architecture group in Cupertino is seeking an engineer to map ML workloads to Apple silicon. This onsite role collaborates with cross-team engineers to build optimized ML workload implementations and to develop performance models and simulations that guide architecture decisions.

You will communicate findings to software and algorithm teams to validate trade-offs and improve execution efficiency, and you will prototype algorithms on CPU, GPU, and Neural Engine.