GPU Performance Engineer, Platform Architecture

Apple • San Diego, CA

1d

About The Position

Imagine what you could do here! At Apple, new ideas have a way of becoming
extraordinary products, services, and customer experiences very quickly. Bring
passion and dedication to your job and there's no telling what you could
accomplish. Dynamic, hard-working people and inspiring, innovative technologies
are the norm here. The people who work here have reinvented entire industries
with all Apple Hardware products. Join us to help deliver the next
groundbreaking Apple product! We are seeking a highly motivated, innovative, and
dedicated engineer to join the Platform Architecture GPU Modeling Team; we are a
group that is driving advanced exploration for next generation GPU architectures
in iPhone, iPad and Mac products. We rely on strong analytical skills and close
collaboration to deliver the best overall solution to our customers. In this
highly visible role, you will be at the center of a chip design effort working
with all disciplines, with a critical impact on getting highly performant
products to millions of customers quickly. This is an opportunity to join
Apple’s world-class GPU team to collaborate, develop and improve GPU simulators
for research, performance analysis, and architectural tuning.

DESCRIPTION

As part of the Platform Architecture GPU Modeling team, you will be working with
architecture, design and software partners on the definition and tuning of
Apple’s future GPU architectures. The role involves the creation and maintenance
of a high-performance C++ model of our next-generation GPUs. Core
responsibilities include coding and debugging a performance and functional model
of the GPU; implementing modern hardware features such as machine learning for
AI, ray tracing, mesh shading, and more. Other responsibilities include working
with the team to ensure model accuracy, feature validation and correlating
against other models and RTL. You’ll develop and utilize diverse tools for
analyzing the substantial amount of data generated by the models. The team is in
a highly collaborative work environment, and you’ll be working with experienced
architects, modeling engineers, compiler engineers, and software teams to assess
the feasibility of new hardware and software features, running performance
experiments, analyzing results, and proposing architectural changes.

Requirements

• Bachelor’s degree
• Experience writing and debugging C++ code
• Experience with scripting languages such as Python or Ruby

Nice To Haves

• 3+ years of experience modeling GPUs, CPUs, or similar
• MS or PhD degree in related field
• Experience coding components such as processing cores, texturing units, caches, memory hierarchies, etc.
• Understanding of GPU/CPU architectures
• Experience debugging performance issues and correlating multiple models
• Understanding of data analysis tools such as: Tableau, pandas, Excel, matplotlib, etc.
• Experience working with cutting edge machine learning and/or graphics applications/games/benchmarks
• Troubleshooting skills

Responsibilities

• coding and debugging a performance and functional model of the GPU
• implementing modern hardware features such as machine learning for AI, ray tracing, mesh shading, and more
• working with the team to ensure model accuracy, feature validation and correlating against other models and RTL
• develop and utilize diverse tools for analyzing the substantial amount of data generated by the models
• working with experienced architects, modeling engineers, compiler engineers, and software teams to assess the feasibility of new hardware and software features, running performance experiments, analyzing results, and proposing architectural changes

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level
• Number of Employees: 5,001-10,000 employees