Staff Engineer AI/HPC Infrastructure
$153,600 - $261,100 USD yearly
Originally posted 11 November 2026 by the employer.
About the role
The Staff Engineer AI/HPC Infrastructure will modify and operate GPU computing platforms for AI training, inference, and memory-system research. The role focuses on hardware/software co-design and real-world AI workload analysis to guide development of HBF, SSD, and memory-centric AI systems.
What you'll do
- Build and manage AI testbeds supporting 8-100+ GPU environments, including server, networking, storage, and software infrastructure.
- Install and optimize LLM serving frameworks such as vLLM, NVIDIA Dynamo, TensorRT-LLM, Triton Inference Server, and SGLang for latency, throughput, and GPU utilization.
- Capture workload traces from training and inference platforms to analyze memory bandwidth, KV-cache growth, storage access patterns, and network traffic.
- Evaluate system bottlenecks across SSD, networking, and GPU interconnects while studying disaggregated memory and storage approaches.
- Build simulation environments and digital-twin platforms based on real workload traces to model future AI infrastructure and evaluate HBM, SSD, and memory products.
- Collaborate with universities, startups, national laboratories, cloud service providers, and OEMs to evaluate emerging architectures and contribute to technical papers and standards.
What you'll need
- Bachelor's degree with 10+ years of experience (including 5+ years of relevant AI experience), Master's with 6+ years (3+ years relevant AI), or PhD with 3+ years (2+ years relevant AI plus coursework).
- Expertise in AI infrastructure engineering and GPU cluster administration using Linux, Kubernetes, Docker, Slurm, and Python.
- Hands-on experience with LLM serving engines and frameworks including vLLM, NVIDIA Dynamo, TensorRT-LLM, Triton, and SGLang.
- Solid background in networking and scale-out systems including InfiniBand, RoCE, RDMA, NCCL, and multi-node inference across distributed AI clusters.
- Technical knowledge of AI accelerator architecture, memory hierarchy optimization, SSD architecture, storage-memory co-design, and hardware/software co-design.
Skills: vLLM, TensorRT-LLM, GPU cluster, InfiniBand, HBM
More semiconductor jobs: SanDisk jobs · California jobs
First seen in our index 0 days ago, well below the 56-day median for Infrastructure/Platform Software roles.
Infrastructure/Platform Software · Infrastructure Platform
|
Open roles in category
1034
|
Median days since first seen
56 d
|
Category median salary (all employers)
$236k
|
See the full market breakdown ▾Category comparison, skills in demand, and who else is hiring
| Metric | SanDisk | All employers we track in this specialty (1034 roles · 94 employers) |
|---|---|---|
| Open roles in this specialty | 8 | 1034 |
| Open roles in the wider Software, Firmware & Systems family | 66 | 6577 · 158 employers |
| Median days since first seen | 12 d | 56 d (+44 d vs this employer) |
| Median salary (USD postings) | n/a | $236k |
Skills observed across this category: vLLM, TensorRT-LLM, GPU cluster, InfiniBand, HBM
Who's hiring in this category
- NVIDIA · 345 open roles · median 60 d
- Qualcomm · 56 open roles · median 61 d
- AMD · 54 open roles · median 36 d
- Graphcore · 47 open roles · median 117 d
- Arm Holdings · 37 open roles · median 14 d
- Intel Corporation · 36 open roles · median 22 d
How we counted: 1034 open Infrastructure/Platform Software (Infrastructure Platform) roles from 94 employers tracked in the SemiconductorJobs index, counted 11 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, and is counted in the wider Software, Firmware & Systems family row. Days are counted from when our index first saw each posting, not from the employer's original posting date.