Senior AI Accelerator Architect — Performance

Cerebras

Sunnyvale (CA)

On-site

USD 175,000 - 275,000

Full time

3 days ago

Be an early applicant

Application generator

Turn this role into an interview — a resume and cover letter built around what this employer wants.

Get past ATS filters

Job summary

Cerebras Systems is seeking a Senior Staff AI Accelerator Performance Architect to guide the evolution of next-generation AI systems. You will connect real workloads to architectural behavior, identify bottlenecks, quantify improvements, and influence roadmaps through rigorous performance analysis.

You will work across architecture, compiler, kernel, runtime and systems teams to validate models with RTL, emulation and silicon measurements, and to shape workload projections and performance

Qualifications

10+ years in performance analysis or architecture exploration in HPC/AI accelerators.
Strong understanding of hardware architecture through multiple domains.
Experience building analytical, simulation-based or trace-driven models (Python/C++).
Knowledge of memory systems, interconnects and parallel execution.

Responsibilities

Own and evolve performance models for next-gen accelerator architectures.
Analyze AI workloads to locate latency, bandwidth, compute, and capacity bottlenecks.
Quantify opportunities to improve latency, throughput and energy efficiency.
Collaborate with architecture, compiler, kernel and systems teams on mappings and optimizations.
Translate complex results into actionable architectural recommendations.

Skills

Performance modeling

Hardware architecture

Python

C++

End-to-end perf analysis

Communications of findings

Education

MS or PhD in EE/CE/CS

Tools

Python

C++

RTL/Emulation

Job description

You will work across architecture, compiler, kernel, runtime and systems teams to validate models with RTL, emulation and silicon measurements, and to shape workload projections and performance

Get your free, confidential resume review.

or drag and drop your file here.