This role pays more than
78%
of similar roles. Most pay
$162,000–$241,750
— the shaded band above.
At the midpoint, this role pays about
$242k
versus about
$202k
for comparable roles.
Based on 240 similar postings.
Employer
About Apple Inc
Apple Inc. is a multinational technology company known for designing and manufacturing consumer electronics, software, and online services, including the iPhone, Mac, iPad, and App Store. Industry: Consumer Electronics & Software
Apple Inc currently has
3587 open roles
on FindRole.
Listed pay typically runs
$165,800–$277,600
across 2768 roles with salary data.
The SW Optimization Engineer AI/ML joins the Platform Architecture group to integrate hardware and software into a unified system. This role focuses on identifying performance bottlenecks in hardware and software while providing recommendations for improvements across Apple Silicon platforms. You will conduct performance studies to validate architecture decisions, create optimized implementations of machine learning workloads for the Neural Engine, GPU, and CPU, and develop performance models for emerging AI/ML techniques. Collaborating with various teams, you will deliver performance analysis instruments and optimized libraries for AI/ML applications. Required skills include experience in software development for hardware IPs like GPUs or accelerators, proficiency in C/C++ and Python, and knowledge of PyTorch, operating system internals, and compiler technologies. The role addresses the technical challenge of optimizing complex machine learning workloads on specialized silicon architectures.
Analyze new and existing workloads to identify performance bottlenecks in hardware and software.
Create optimized implementations of machine learning workloads on CPU, GPU, and Neural Engine.
Conduct performance studies to inform and validate architectural decisions for Apple Silicon.
Develop performance models for emerging AI/ML techniques and analyze system architecture trade-offs.
Deliver performance analysis instruments, optimized libraries, and frameworks for AI/ML applications.
Provide feedback to silicon and software teams based on identified patterns of performance challenges.
Prototype and benchmark algorithms while creating high-level complexity models.
Develop targeted, highly efficient low-level performance libraries for AI/ML accelerators and GPUs.
What we're looking for
Bachelor's degree or equivalent experience in Computer Engineering, Computer Science, or a related field.
Experience in software development for AI/ML hardware accelerators, GPUs, image/video encoders, or similar.
Experience with AI/ML, graphics, or HPC performance benchmarks and workloads.
M.S. or Ph.D. in Computer Science, Computer Engineering, Electrical Engineering, or a closely related field (preferred).
10+ years of experience in software performance optimization, analysis tools, processes, and developing efficient computational algorithms (preferred).
Proficiency in C/C++ programming languages and scripting languages like Python (preferred).
Ability to prototype, benchmark, and analyze performance metrics on CPU, GPU, and Neural Engine platforms (preferred).
Knowledge of computer architecture, operating system internals, compiler technologies, and AI/ML frameworks like PyTorch (preferred).
System Power Modeling
AI Systems Architecture
C++
Python
Perl
Ruby
Scripting
Platform Management Framework
Power Analysis
Thermal Constraints
Oscilloscopes
Logic Analyzers
Validation
Automation
Hardware Modeling