Senior AI Inference Performance Engineer

Cerebras Systems, Inc.; Sunnyvale (CA); On-site; USD 130,000 - 160,000; Full time

Benefits offered by this job

Opportunities for contributing to open-source projects

Stable work environment with startup vitality

Job summary

Cerebras Systems, Inc. is looking for a Senior Performance Engineer to enhance the performance benchmarking and competitive pricing models for their AI chip. The ideal candidate will have extensive experience with open-source inference frameworks and an understanding of ML systems.

This role critically combines technical acumen and market analysis to ensure Cerebras maintains a compelling value proposition in the AI landscape. Join a team committed to transforming AI with innovative architecture.

Qualifications

• 5+ years of experience in ML systems or ML research engineering.
• Deep practical experience with open-source inference frameworks.
• Strong understanding of tokens, throughput, latency, and precision trade-offs.

Responsibilities

• Design benchmark suites for inference workloads.
• Stay updated with GPU optimization communities.
• Build competitive pricing models across major inference providers.
• Monitor industry announcements and synthesize findings.
• Collaborate with Sales for competitive analyses.

Skills

Experience with open-source inference frameworks (vLLM, SGLang, TensorRT-LLM)

ML systems or high-performance computing experience

Understanding of LLM inference economics

Knowledge of transformer model architectures

Tools

CUDA

Triton

Job description

Cerebras Systems, Inc. is looking for a Senior Performance Engineer to enhance the performance benchmarking and competitive pricing models for their AI chip. The ideal candidate will have extensive experience with open-source inference frameworks and an understanding of ML systems.

This role critically combines technical acumen and market analysis to ensure Cerebras maintains a compelling value proposition in the AI landscape. Join a team committed to transforming AI with innovative architecture.