Software Engineer Host and Network IO
Cerebras builds wafer-scale AI computing systems and a cloud inference platform for training, fine-tuning, and serving AI models.
About Cerebras Systems, Inc.
Cerebras Systems is an AI-infrastructure company founded in 2015. It sells rack-scale wafer-scale computing systems and provides cloud-based, API-accessible AI inference alongside on-premises deployments.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will develop software that exposes hardware IO capabilities and manage a shared IO API. You will build hardware-control systems, debug AI-cluster network performance, analyze packet traces, and create telemetry tools. You will optimize resource use, integrate networking technologies, and lead cross-functional network IO projects.
Requirements
- Master's or PhD in Computer Science or Electrical Engineering with 1 year of industry experience, or 3+ years of industry experience
- Experience in large software environments
- Embedded systems, hardware-software co-design, and driver development
- TCP and RoCE
- Network debugging tools such as Wireshark, or willingness to learn
- Network switch environment familiarity with platforms such as Arista or Juniper, or willingness to learn
Responsibilities
- Develop x86 and ARM software for hardware IO capabilities
- Govern a generic IO API
- Develop hardware control and configuration subsystems
- Drive network-performance debugging for AI clusters
- Analyze network statistics and packet traces to resolve bottlenecks
- Develop network and IO telemetry tools
- Optimize CPU and memory use with kernel-bypass and zero-copy techniques
- Integrate networking technologies and protocols
- Lead cross-functional network IO projects