GPU Performance Architect

Amd

Confirmed live 2 days ago High trust
Hybrid

Quick summary

Work type
Hybrid
Location
Folsom, CASanta Clara, CA
Salary
$144,800–$217,200 / yr
Posted
67 days ago
Freshness
Confirmed live 2 days ago
Closes
Jul 6, 2027

Market check

Salary context

Below market

How this pay compares to similar roles

Similar $219k
This role $181k
$129k most similar roles pay here $290k

This role pays less than 79% of similar roles. Most pay $187,850–$250,000 — the shaded band above. At the midpoint, this role pays about $181k versus about $219k for comparable roles.

Based on 240 similar postings.

Employer

About Amd

AMD (Advanced Micro Devices) is a semiconductor company that develops high-performance processors, graphics cards, and adaptive computing solutions for gaming, data centers, and embedded markets. Industry: Semiconductors

Amd currently has 367 open roles on FindRole.

Listed pay typically runs $166,400–$249,600 across 367 roles with salary data.

Most-posted roles

View all roles at Amd

At a glance

TL;DR · GPU Performance Architect

As a GPU Performance Architect within the RTG Architecture team, you will collaborate with architecture, software, and design teams to develop next-generation products for data centers and supercomputers. Your daily responsibilities include developing high-level and detailed performance models for GPU subsystems, SoCs, and multi-GPU systems to predict application performance and establish key metrics like latency and throughput. You will perform kernel prototyping across the stack from firmware to architectural models, write and optimize GPU kernels using assembly, Hip, and Triton, and develop microbenchmarks for competitive analysis. The role requires expertise in C, C++, Python, and machine learning frameworks like TensorFlow and PyTorch. You will address technical challenges involving ML/HPC workloads, Network-on-Chip design, and the compiler stack including MLIR and LLVM to provide insights into emerging hardware and software technologies.

What you'll do

  • Develop high-level and detailed performance models for GPU subsystems, SoCs, and multi-GPU systems.
  • Build analytical, simulation-based, and workload-driven models to predict application performance.
  • Generate performance projections for future architectures and product configurations.
  • Establish performance, latency, throughput, and efficiency KPIs during early product definition.
  • Prototype ideas across the stack from software and firmware to architectural models to validate new technologies.
  • Write and optimize GPU kernels at various levels of abstraction including assembly, Hip, and Triton.
  • Create microbenchmarks for competitive analysis and performance verification.
  • Collect and summarize data and simulation results for use by architects and design teams.

What we're looking for

  • Candidates must possess a Bachelor's, Master's, or PhD degree in Electrical Engineering, Computer Science, or Computer Engineering.
  • Experience developing high-level and cycle-accurate performance simulators is preferred.
  • Knowledge of GPU architectures and basic knowledge of CPU architecture is required.
  • Proficiency in programming languages including C, C++, and scripting languages like Python is required.
  • Experience with machine learning frameworks such as TensorFlow and PyTorch is preferred.
  • Understanding of Compute APIs such as CUDA or OpenCL is preferred.
  • Experience building libraries for key ML operators like GEMMs, Attention, and Collectives is preferred.
  • Experience in the compiler stack (MLIR, LLVM) or hardware modeling (RTL, SystemC) is a plus.

More like this

Similar roles