Principal Software Developer, GPU AI/HPC Kernels

Amd

Hybrid

Quick summary

Work type
Hybrid
Location
Austin, TX
Posted
120 days ago
Closes
Feb 27, 2027

Market check

Salary context

How this pay compares to similar roles

Similar $206k
$140k most similar roles pay here $265k

This listing doesn't post a salary. Most similar roles pay $177,250–$235,750.

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 56 open roles on FindRole.

Most-posted roles

View all roles at Amd

At a glance

TL;DR · Principal Software Developer, GPU AI/HPC Kernels

AMD seeks a principal software developer to join its ROCm GPU-compute mathematical libraries team in Austin, Texas. As a key contributor, you will develop and optimize GPU computational software libraries for AI and HPC applications using C++, Python, HIP, and assembly languages. You’ll collaborate with management to plan and deliver cutting-edge software solutions, supervise small development teams, and perform performance optimizations to maximize hardware utilization. Ideal candidates have over 10 years of professional experience in software development, proficiency in C/C++ and Python, and expertise in GPU software validation involving HIP or similar technologies. They should also possess strong problem-solving skills and excellent communication abilities, with a preference for advanced degrees in Computer Science or related fields. This role involves reimagining the end-to-end software libraries lifecycle to create unified software fabrics across multiple ASICs, bridging hardware and software ecosystems to ensure scalable solutions for emerging platforms.

What you'll do

  • Develop software in C++, Python, HIP, and other technologies for GPU mathematical operations.
  • Design GPU computational libraries for AI and HPC applications.
  • Conduct performance optimizations and projections to maximize hardware utilization.
  • Identify and resolve quality issues with close collaboration with development teams.
  • Provide technical leadership for GPU architecture decisions impacting multi-ASIC platforms.
  • Define and lead bring-up strategies for new GPU platforms across multiple ASICs.
  • Drive hardware/software interface definition to ensure alignment with architectural intent.

What we're looking for

  • 10+ years of professional software development experience
  • Expertise in C/C++, Python, and GPU programming technologies like HIP, CUDA
  • Proven ability to lead and manage junior to mid-level developers
  • Experience with software performance optimization and debugging
  • Strong background in designing and developing software libraries and APIs
  • Academic credentials: Master’s or Doctoral degree in a relevant field
  • Hands-on experience in hardware/software co-design and cross-layer debugging

More like this

Similar roles

Software Development Engineer, GPU

Amd

Canada 96 days ago
C++ HLSL GLSL Shader Real-time rendering Graphics algorithms AAA game development Project management Technical leadership Debugging Optimization Documentation Communication skills Problem-solving skills Planning and scheduling Middleware development English communication

System Software Engineer, GPU Development Tools

Nvidia

Santa Clara, CA +1 83 days ago $124,000$195,500
C++ Python CUDA OpenGL Vulkan Docker Kubernetes CI/CD Git Linux Virtual Machines Containers Prometheus Grafana PostgreSQL AWS Azure Google Cloud Platform

GPU AI Compiler Engineer

Qualcomm

San Diego, CA 58 days ago $141,600$212,400
C/C++ LLVM SYCL CUDA OpenCL MLIR GPU architecture Machine Learning Graph Compiler Data structures Algorithms Object-oriented programming