Principal Network Architect: Low-Latency AI Interconnects

Cerebras; Sunnyvale (CA); On-site; USD 250,000 - 360,000; Full time

Job summary

Cerebras Systems is seeking a Principal Network Systems Architect to define scale-out and, especially, scale-up network architectures for Cerebras platforms, including proprietary interconnects and switching. You will translate workload patterns into fabric requirements and guide the design through implementation and qualification.

Your work focuses on low-latency scale-up networks with breadth across scale-out and customer-facing networks, deciding when standards-based or custom protocols are

Qualifications

• BS, MS, or PhD in Electrical Engineering, Computer Engineering, Computer Science, or a related field, or equivalent practical experience.
• Typically 12+ years of relevant industry experience with principal-level ownership of a networking or HPC system.
• Deep expertise in low-latency interconnects, fabric protocols, switch architecture, and data-plane depth.
• Experience deriving network requirements from incomplete workload and system information across software, accelerator I/O, topology, and physical constraints.
• Working knowledge of Ethernet, IP, and RDMA; familiarity with BGP and EVPN and tradeoffs between routed networks and low-latency designs.
• Architectural judgment grounded in prior implementation, modeling, silicon, lab bring-up, or debugging work.
• Record of making clear technical decisions and influencing engineering leaders, teams, suppliers, and customers.

Responsibilities

• Set the multi-generation architecture and roadmap for Cerebras scale-up networks and interfaces to scale-out networks.
• Work with multiple teams to understand mapping and communication choices, deriving bandwidth, latency, ordering, availability, and serviceability.
• Define fabric topology, protocols, and switch behavior, including routing, buffering, flow control, reliability, and fault containment.
• Decide when to use standards-based technology or merchant silicon vs. custom protocol or switching design.
• Use performance models, traffic simulation, prototypes, and lab data to test architecture choices and set acceptance criteria.
• Write architecture and interface specifications, lead design reviews, and drive cross-layer decisions through implementation, bring-up, and qualification.

Skills

Low-latency design

Architectural ownership

System-level thinking

Cross-team collaboration

Workload mapping

Data-plane depth

Decision making

Education

BS/MS/PhD in EE/CE/CS or equivalent

Tools

NCCL

RCCL

MPI

SHMEM

RDMA

Job description

Cerebras Systems is seeking a Principal Network Systems Architect to define scale-out and, especially, scale-up network architectures for Cerebras platforms, including proprietary interconnects and switching. You will translate workload patterns into fabric requirements and guide the design through implementation and qualification.

Your work focuses on low-latency scale-up networks with breadth across scale-out and customer-facing networks, deciding when standards-based or custom protocols are