Principal AI Software & Systems Architect - TX, Austin, United States
Originally posted 26 May 2026 by the employer.
About the role
AMD is seeking a Principal AI Software & Systems Architect to drive adoption of AMD AI solutions across on-premise and enterprise environments. The role focuses on mapping real-world AI workloads across CPUs, GPUs, NPUs, and distributed infrastructure to optimize performance and total cost of ownership.
What you'll do
- Analyze customer and ISV AI deployments, including models, frameworks, runtimes, serving infrastructure, and dependencies. - Evaluate workload partitioning across CPUs, GPUs, NPUs, and distributed compute resources for performance and total cost of ownership (TCO) modeling. - Identify software, runtime, framework, compiler, driver, and hardware dependencies impacting deployment on AMD platforms. - Analyze workload performance and identify bottlenecks related to compute, memory, communication, storage, power, and utilization. - Serve as technical lead for strategic customer and ISV engagements, conducting architecture reviews and guiding workload migration. - Translate real-world workload requirements into software, platform, and hardware roadmap requirements.
What you'll need
- Bachelor’s or Master's degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent. - Deep understanding of AI workload architecture and hardware-aware software design. - Experience analyzing complex software stacks spanning frameworks, runtimes, drivers, operating systems, and hardware. - Experience analyzing performance and executing technical work to map AI workloads onto accelerator architectures. - Experience with customer and ISV technical engagement and workload migration.
Nice to have
- Deep understanding of AI inference systems, model serving architectures, AI infrastructure, and performance optimization in production environments. - Strong understanding of AI software stacks, including frameworks, runtimes, compilers, drivers, and accelerator libraries. - Experience analyzing and optimizing software across multiple layers of the stack, from application and framework behavior through runtime, operating system, driver, and hardware interactions. - Strong understanding of heterogeneous computing architectures and workload partitioning across CPUs, GPUs, NPUs, and distributed compute resources. - Experience evaluating workload characteristics such as latency, throughput, memory footprint, communication patterns, scalability, power efficiency, and total cost of ownership (TCO). - Experience enabling, optimizing, porting, or debugging software on GPU or other accelerator platforms.
Skills: ROCm, AI inferencing, GPU computing, enterprise software, Python, C/C++
This role has been open 150 days — well above the 55-day median for Infrastructure/Platform Software roles.
Infrastructure/Platform Software · GPU Software Stack
|
Open roles in category
242
|
Median days open
55 d
|
Median salary
$220k
|
See the full market breakdown ▾Category comparison, skills in demand, and who else is hiring
| Metric | AMD | All employers we track in this specialty (242 roles · 20 employers) |
|---|---|---|
| Open roles in this specialty | 51 | 242 |
| Open roles in the wider Software, Firmware & Systems family | 389 | 6277 · 144 employers |
| Median days open | 53 d | 55 d (−2 d vs this employer) |
| Median salary (USD postings) | — | $220k |
Skills observed across this category: ROCm, AI inferencing, GPU computing, enterprise software, Python, C/C++
Who's hiring in this category
- NVIDIA · 120 open roles · median 57 d
- AMD (this employer) · 52 open roles · median 53 d
- Intel Corporation · 25 open roles · median 14 d
- Qualcomm · 18 open roles · median 128 d
- Arm Holdings · 5 open roles · median 121 d
- Graphcore · 4 open roles · median 3 d
How we counted: 242 open Infrastructure/Platform Software (GPU Software Stack) roles from 20 employers tracked in the SemiconductorJobs index, counted 24 Oct 2026. Specialty figures count only roles carrying this exact specialty label, so an employer's related work in neighbouring specialties is not included there — it is counted in the wider Software, Firmware & Systems family row. Figures refresh nightly.