Startups · AI

Network Systems Architect

Cerebras Systems · Sunnyvale, CA · Remote

← All jobs
About Cerebras Systems

Cerebras powers the world's fastest AI inference on the biggest wafer chip. Cerebras CS-4 delivers up to 30x faster inference than GPUs.

About the role

As a Network Systems Architect, you will define the scale out, and particularly scale-up network architecture for current and future Cerebras platforms, including proprietary accelerator interconnects, protocols, and switching. Requirements will not arrive as a finished bandwidth and latency specification.

What they're looking for

  • Relevant experience may include one or more of the following
  • • Proprietary accelerator interconnects or scale-up technologies such as NVLink, xGMI, TPU ICI, Xe Link, or UALink-class systems
  • • Switch ASIC, NIC or DPU, accelerator I/O, transport offload, collective acceleration, coherent memory, or custom-fabric work
  • • FPGA architecture or mapping experience, or work with other placement-sensitive systems where topology materially affects communication
  • • AI or HPC communication stacks such as NCCL, RCCL, MPI, SHMEM, or proprietary collective libraries
  • • RoCE, InfiniBand, PCIe, CXL, or other relevant scale-out and I/O technologies
More about this role

Cerebras Systems builds the world's largest AI chip, 56 times larger than GPUs. This architecture allows Cerebras to deliver industry-leading training and inference speeds; over 10 times faster than GPU-based hyperscale cloud inference services.

This order of magnitude increase in speed is transforming the user experience of AI applications, unlocking real-time iteration and increasing intelligence via additional agentic computation.

Cerebras works with the leading model labs, global enterprises, and cutting-edge AI-native startups. OpenAI recently announced a multi-year partnership with Cerebras, to deploy 750 megawatts of scale, transforming key workloads with ultra high-speed inference.

As a Network Systems Architect, you will define the scale out, and particularly scale-up network architecture for current and future Cerebras platforms, including proprietary accelerator interconnects, protocols, and switching. Requirements will not arrive as a finished bandwidth and latency specification. Working with application, compiler, runtime, and systems teams, you will study communication patterns, workload partitioning and placement, data and memory movement, synchronization, locality,...

Read the full posting on Cerebras Systems's site ↗

Software Engineering

Build your edge while you search

Free tools for founders and investors, plus VC Unfiltered, our take on startups, venture and the people who build them.