AI Cluster Network IO Engineer for Low-Latency Throughput

Cerebras; United States; Remote; USD 150,000 - 210,000; Full time

Job summary

Cerebras Systems seeks a software developer for the Host and Network IO team to optimize bandwidth and latency on a distributed AI/HPC system. You will collaborate with AI application teams, cluster architects, and FPGA/ASIC groups to deploy robust IO solutions and reduce congestion.

The role demands proficiency in socket programming, RDMA Verbs, and performance tuning to improve real-world AI throughput. You will lead network debugging across large clusters and help shape architectural changes

Qualifications

• Master's/PhD in Computer Science or Electrical Engineering with 1 year industry experience, or 3+ years industry experience.
• Experience in large software environments.
• Embedded systems, HW/SW co-design, and driver development.
• Networking experience.

Responsibilities

• Develop x86/ARM software to expose IO capabilities for AI/HPC teams.
• Govern a generic IO API with multiple internal users.
• Develop control and configuration subsystems interacting with Cerebras hardware.
• Drive network performance debugging of large AI clusters.
• Collect and analyze network statistics and traces to diagnose bottlenecks.
• Develop telemetry tools to increase visibility into the network and IO datapath.
• Optimize CPU/memory utilization with kernel bypass and zero-copy techniques.
• Integrate advanced networking technologies and protocols.
• Lead cross-functional technical projects across teams to improve IO solutions.
• Foster clear communication across teams and stakeholders.

Skills

Advanced degree in CS/EE + 1yr or 3+yr

Experience in large software envs

Embedded systems HW/SW co-design

Network experience

Education

Master's/PhD in Computer Science or Electrical Engineering

Job description

Cerebras Systems seeks a software developer for the Host and Network IO team to optimize bandwidth and latency on a distributed AI/HPC system. You will collaborate with AI application teams, cluster architects, and FPGA/ASIC groups to deploy robust IO solutions and reduce congestion.

The role demands proficiency in socket programming, RDMA Verbs, and performance tuning to improve real-world AI throughput. You will lead network debugging across large clusters and help shape architectural changes