Principal AI Systems Performance Engineer
Originally posted 2 November 2026 by the employer.
About the role
As a Principal AI Systems Performance Engineer, you will direct experiments on AI training, inference, and agentic workloads across complete server systems. Your work will turn system-level measurements across compute, memory, interconnects, and storage into product requirements for future DRAM, HBM, CXL, and storage products.
What you'll do
- Develop and perform controlled, representative AI experiments on current and pre-production CPU-GPU systems, evaluating memory and storage performance, power, and thermal tradeoffs.
- Profile workloads from GPU memory through system memory, CXL-attached memory, storage, and interconnects, evaluating node to cluster scale performance, efficiency, latency, and scalability.
- Evaluate data placement, migration, prefetching, compression, and tiering approaches for KV cache, model state, and other AI data.
- Build and validate performance models, simulations, traces, and reusable benchmarking tools to forecast future system requirements.
- Lead cross-functional projects and transform outcomes into product requirements, develop recommendations, publications, demonstrations, and customer mentorship.
What you'll need
- Bachelor’s degree with at least 10 years of experience, Master’s degree with 5 or more years, or a Ph.D. with demonstrated experience in a related field.
- Hands-on experience profiling and benchmarking multi-GPU AI systems across single-node and multi-node environments.
- Understanding of transformer workloads, memory behavior, distributed systems, and agentic AI infrastructure.
- Experience with PyTorch, an AI serving platform, Python, and at least one of C, C++, CUDA, or Triton.
- Ability to lead controlled experiments and communicate results to technical and business audiences.
Nice to have
- Experience with memory tiering, KV-cache management, or emerging AI workloads.
- Experience with disaggregated infrastructure or performance modeling and simulation.
- Relevant research, patents, open-source contributions, or other substantial technical work.
Skills: CUDA, PyTorch, performance modeling, GPU memory, benchmarking, Triton
More semiconductor jobs: Micron Technology jobs · Austin jobs
This role has been open 0 days — well below the 58-day median for Infrastructure/Platform Software roles.
Infrastructure/Platform Software · GPU Software Stack
|
Open roles in category
277
|
Median days open
58 d
|
Category median salary (all employers)
$225k
|
See the full market breakdown ▾Category comparison, skills in demand, and who else is hiring
| Metric | Micron Technology | All employers we track in this specialty (277 roles · 19 employers) |
|---|---|---|
| Open roles in this specialty | 1 | 277 |
| Open roles in the wider Software, Firmware & Systems family | 240 | 6322 · 144 employers |
| Median days open | 0 d | 58 d (−58 d vs this employer) |
| Median salary (USD postings) | — | $225k |
Skills observed across this category: CUDA, PyTorch, performance modeling, GPU memory, benchmarking, Triton
Who's hiring in this category
- NVIDIA · 143 open roles · median 66 d
- AMD · 52 open roles · median 61 d
- Intel Corporation · 31 open roles · median 17 d
- Qualcomm · 17 open roles · median 192 d
- Arm Holdings · 10 open roles · median 5 d
- Graphcore · 4 open roles · median 12 d
How we counted: 277 open Infrastructure/Platform Software (GPU Software Stack) roles from 19 employers tracked in the SemiconductorJobs index, counted 2 Nov 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.