Software Engineer - Host and Network IO
Cerebras · Sunnyvale, CA
About this role
## Software Engineer - Host and Network IO ### About the Role Cerebras Systems builds the world’s largest AI chip—56x larger than GPUs—enabling industry-leading training and inference speeds. The **Host and Network IO Team** develops the full IO path between distributed server nodes, through the cluster, down to the custom **RoCE network stack**, and across proprietary IOs onto the **WSE**. As a software developer on this team, you’ll interface with AI application-level IO teams, cluster architecture teams, and FPGA/ASIC teams to optimize **bandwidth and latency** while minimizing congestion, pauses, pause spreading, and unfairness. You’ll use strong **socket programming** for robust management operations and **RDMA Verbs** to optimize CPU resources and shape network traffic. You’ll also lead network debug, analyze network behavior, and develop tools/telemetry to improve visibility into the network and IO datapath. ### 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 - Build control and configuration subsystems interacting directly with Cerebras hardware - Drive network performance debugging for large AI clusters - Gather and analyze network statistics and packet traces to root cause bottlenecks - Develop tools and telemetry for increased network/IO visibility - Optimize CPU/memory utilization using **kernel bypass** and **zero-copy** techniques - Integrate leading-edge networking technologies and protocols - Lead cross-functional technical projects across multiple software and hardware teams - Communicate clearly and effectively with teams and stakeholders ### Skills & Qualifications - **Master’s/PhD** in Computer Science or Electrical Engineering + **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 network protocols (e.g., **TCP, RoCE**) and network debugging tools (e.g., **Wireshark**) or willingness to learn - Some familiarity with network switch environments (e.g., **Arista, Juniper**) or willingness to learn - Detail-oriented with the ability to zoom out and embrace new challenges ### Why Join Cerebras People who are serious about software make their own hardware. Cerebras is building a breakthrough AI platform beyond GPU constraints, publishing/open-sourcing cutting-edge research, and working on one of the fastest AI supercomputers in the world—backed by startup vitality and a simple, non-corporate culture. *Apply today and help drive groundbreaking advancements in AI.* *Equal Opportunity Employer.*
Listing freshness
CronJobs last confirmed this listing 1h ago. If its source stops confirming the opening for seven days, this page is removed from active inventory.