Principal Cloud & AI Workload Performance Analysis Engineer - TX, Home Office, United States
Originally posted 11 September 2026 by the employer.
About the role
This role involves owning workloads exposed to hyperscaler custom CPUs, Arm ecosystem momentum, and AI-system architecture, measuring and explaining cloud-native performance, price-performance, performance-per-watt, software maturity, and host-CPU effects on accelerator utilization. The work requires comparing AMD EPYC with Intel, Arm Neoverse-based platforms, cloud-custom CPUs, NVIDIA Grace-class systems, and other emerging designs.
What you'll do
- Define and maintain the cloud-native and AI workload taxonomy used in competitive analysis and forecasting.
- Design reproducible methods for web services, microservices, containers, scale-out data processing, caching, cloud infrastructure, and CPU-support functions in AI systems.
- Compare AMD, Intel, Arm, and cloud-custom environments using aligned software versions, tuning policies, instance shapes, and service-level objectives.
- Analyze host-CPU impact on accelerator utilization, input pipelines, communication overheads, memory movement, NUMA behavior, and end-to-end AI system throughput.
- Measure and model performance-per-watt, density, utilization, and price-performance where inputs can be normalized defensibly.
- Track compiler, kernel, library, orchestration, and migration factors that affect Arm and cross-ISA deployability.
What you'll need
- Strong Linux systems skills and proficiency with automation or scripting for deployment and analysis.
- Experience with containers, orchestration, and cloud-native workloads at meaningful scale.
- Ability to design fair experiments across dissimilar CPU architectures and platform configurations.
- Experience with application profiling, hardware performance counters, and system telemetry.
- Understanding of AI host-side bottlenecks, CPU-to-accelerator data movement, NUMA, memory bandwidth, and interconnect effects.
Nice to have
- Hands-on experience with Arm Neoverse or cloud-custom Arm platforms.
- Experience benchmarking major cloud-service-provider instances and services.
- Experience measuring AI system throughput or accelerator utilization as a function of host-CPU and platform behavior.
- Familiarity with LPDDR or HBM-attached CPU systems, coherent CPU-accelerator links, high-speed networking, or storage pipelines.
- Participation in MLCommons, OCP, or other relevant benchmark or standards communities.
Skills: cloud-native performance, AI-system architecture, containers, orchestration, Linux systems, accelerator utilization
This role has been open 33 days — well below the 50-day median for Infrastructure/Platform Software roles.
Infrastructure/Platform Software · Infrastructure Platform
|
Open roles in category
973
|
Median days open
50 d
|
Median salary
$235k
|
See the full market breakdown ▾Category comparison, skills in demand, and who else is hiring
| Metric | AMD | All employers we track in this specialty (973 roles · 85 employers) |
|---|---|---|
| Open roles in this specialty | 55 | 973 |
| Open roles in the wider Software, Firmware & Systems family | 404 | 6145 · 145 employers |
| Median days open | 21 d | 50 d (−29 d vs this employer) |
| Median salary (USD postings) | — | $235k |
Skills observed across this category: cloud-native performance, AI-system architecture, containers, orchestration, Linux systems, accelerator utilization
Who's hiring in this category
- NVIDIA · 289 open roles · median 44 d
- Qualcomm · 61 open roles · median 57 d
- AMD (this employer) · 57 open roles · median 21 d
- Graphcore · 44 open roles · median 90 d
- Intel Corporation · 40 open roles · median 9 d
- Cerebras · 39 open roles · median 80 d
How we counted: 973 open Infrastructure/Platform Software (Infrastructure Platform) roles from 85 employers tracked in the SemiconductorJobs index, counted 15 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.