Cerebras
Compute Server Platform Architect
United States and Canada · Full-time · Posted Feb 18, 2026
- Experience
- PhD: + 8 years industry experience; Master’s/Bachelor’s: + 10 years industry experience. Also requires 5+ years in a relevant field.
- Education
- PhD in Computer Science or Electrical/Computer Engineering; or Master’s/Bachelor’s in CS or EE.
About the job
Own the server-side architecture for Cerebras AI clusters, translating workload needs into platform designs and guiding evaluation, qualification, and production adoption.
What you'll do
- Define server roles, configurations, capacity formulas, and lifecycle strategies for clusters.
- Model and validate performance, translate runtime workloads into hardware requirements, and drive cross-stack fixes.
- Lead vendor evaluations, define platform baselines and qualification criteria, and support bring-up and debugging.
What they ask for
- 5+ years in server platform architecture, systems performance engineering, or large-scale infrastructure design for AI/ML, HPC, or performance-sensitive distributed systems.
- Deep understanding of x86 server architecture and strong Linux systems knowledge.
- Experience with high-performance IO paths, including NIC behavior, RDMA/RoCE, and NVMe.
- Ability to create and empirically validate capacity and performance models; experience working with vendors/partners.
Nice to have
- Familiarity with C, C++, and Python.
Tools
x86, Linux, RDMA/RoCE, NVMe, C, C++, Python, CXL
AI focus
Architect the server platform supporting AI cluster training and inference, including runtime services and performance, scalability, and reliability.
Model-written summary · the original posting is authoritative