Senior Staff AI Accelerator Performance Architect
Cerebras builds wafer-scale AI computing systems and a cloud inference platform for training, fine-tuning, and serving AI models.
About Cerebras Systems, Inc.
Cerebras Systems is an AI-infrastructure company founded in 2015. It sells rack-scale wafer-scale computing systems and provides cloud-based, API-accessible AI inference alongside on-premises deployments.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will develop performance models for accelerator and system architectures, analyze AI workloads, and identify hardware and software bottlenecks. You will evaluate architectural features, validate models against RTL, emulation, and silicon measurements, and translate findings into recommendations for architecture and product decisions.
Requirements
- 7+ years of performance analysis, modeling, or architecture exploration experience
- Hardware architecture knowledge
- Performance modeling with Python, C++, or similar environments
- Knowledge of processor architecture, memory systems, interconnects, and parallel execution
- Workload profiling and quantitative validation
- MS or PhD in Electrical Engineering, Computer Engineering, Computer Science, or equivalent practical experience
Responsibilities
- Develop and evolve performance models and modeling methodologies
- Analyze AI workloads from kernels to end-to-end execution
- Identify hardware and software bottlenecks
- Evaluate architectural features and quantify performance improvements
- Develop workload projections and competitive performance analyses
- Validate models against RTL, emulation, and silicon measurements
- Define representative workloads and performance targets
Benefits
- Bonus
- Equity