Low-Latency AI Network Architect

Cerebras; Sunnyvale (CA); On-site; USD 180,000 - 280,000; Full time

Job summary

Cerebras Systems is seeking a Network Systems Architect in Sunnyvale, CA to design scale-up network fabrics and interconnects for Cerebras platforms. You will study communication patterns and workload placement with cross-functional teams to translate needs into measurable fabric requirements.

The role focuses on low-latency scale-up networks, with some exposure to scale-out and customer-facing networks. Hands-on modeling, bring-up, and debugging underpin architecture decisions.

Qualifications

• Experience with low-latency network architecture and interconnects.
• Ability to translate workload patterns into fabric requirements.
• Hands-on experience with modeling, bring-up, and debugging.
• Ability to write architecture specifications and guide prototypes.

Responsibilities

• Set the multi-generation architecture and roadmap for Cerebras scale-up networks and interfaces.
• Collaborate with application, compiler, runtime, and communication-library teams to determine bandwidth, latency, ordering, availability and serviceability requirements.
• Define fabric topology, protocols, and switch behavior, including routing, buffering, flow control, reliability, and fault containment.
• Decide when standards-based tech or merchant silicon vs. custom protocols/switches is justified.
• 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.

Skills

Low-latency design

Scale-up networking

Performance modeling

Cross-functional collaboration

Prototyping and bring-up

Tools

Simulation tools

Lab measurements

Job description

Cerebras Systems is seeking a Network Systems Architect in Sunnyvale, CA to design scale-up network fabrics and interconnects for Cerebras platforms. You will study communication patterns and workload placement with cross-functional teams to translate needs into measurable fabric requirements.

The role focuses on low-latency scale-up networks, with some exposure to scale-out and customer-facing networks. Hands-on modeling, bring-up, and debugging underpin architecture decisions.