Teams at Cerebras Systems Inc.

Define multi-generation scale-up and scale-out network architectures for AI accelerator platforms. Translate application and system communication patterns into bandwidth, latency, reliability, and serviceability requirements. Specify fabric topologies, protocols, routing, switching, buffering, flow control, and fault containment. Evaluate standards-based versus custom designs using performance models, simulations, prototypes, and lab data. Lead architecture specifications, design reviews, implementation decisions, bring-up, and qualification across hardware and software teams.

Lead performance modeling and architecture analysis for next-generation AI accelerators. Build analytical, simulation-based, and trace-driven models; analyze AI workloads and hardware bottlenecks; evaluate architectural features; validate models against RTL, emulation, and silicon data; and provide recommendations that influence hardware and software roadmaps. Partner with architecture, compiler, kernel, runtime, and systems teams to improve latency, throughput, utilization, and energy efficiency.

Leads end-to-end technical programs for AI compute systems and data center deployments. Owns engineering improvement schedules, hardware change implementation, software integration, validation, risk mitigation, and cost management. Coordinates product, engineering, manufacturing, supply chain, procurement, and software teams while presenting program status and risks to senior leadership. Develops scalable processes and ensures hardware and software deliverables meet performance, budget, and schedule objectives.