Host & Network IO Engineer — High-Performance AI Clusters

Engg; Sunnyvale (CA); On-site; USD 150,000 - 190,000; Full time

Job summary

Cerebras Systems is seeking a software developer for the Host and Network IO Team. You will implement the IO path between a distributed set of server nodes and Cerebras' proprietary ROCE stack, collaborating with AI teams and FPGA/ASIC groups to optimize bandwidth and latency.

You should have strong socket programming skills and RDMA Verbs experience to reduce CPU usage and improve AI performance. This role blends software, hardware, and network work at a fast-moving AI hardware company.

Qualifications

• Master's or PhD in CS or EE plus 1 year industry experience, or 3+ years industry experience.
• Experience in large software environments.
• Embedded systems, HW/SW co-design, and some driver development.
• Familiarity with TCP, RoCE, and network debugging tools.
• Willingness to learn Arista/Juniper network environments.

Responsibilities

• Develop x86 & ARM software to expose next-generation hardware IO capabilities for AI/HPC application teams
• Govern a generic IO API with multiple internal users.
• Develop control and configuration subsystems directly interacting with Cerebras hardware
• Drive network performance debug of large AI clusters
• Gather and analyze network statistics and packet traces to root cause and alleviate bottlenecks and sub-optimalities.
• Develop tools/telemetry for increasing visibility into the network and IO datapath.
• Optimize cpu/mem utilization leveraging kernel bypass and zero-copy techniques
• Integrate leading edge networking technologies and protocols
• Lead cross-functional technical projects spanning multiple teams and integrating diverse software and hardware components to deliver an improved network IO solution.
• Foster clear and effective communication across teams and stakeholders.

Skills

Large software environments

Embedded systems

HW/SW co-design

Driver development

TCP/RoCE networks

Education

Advanced degree in CS or EE

Tools

Wireshark

Job description

Cerebras Systems is seeking a software developer for the Host and Network IO Team. You will implement the IO path between a distributed set of server nodes and Cerebras' proprietary ROCE stack, collaborating with AI teams and FPGA/ASIC groups to optimize bandwidth and latency.

You should have strong socket programming skills and RDMA Verbs experience to reduce CPU usage and improve AI performance. This role blends software, hardware, and network work at a fast-moving AI hardware company.