Cerebras
Engineering Lead, Kernel Reliability
United States and Canada · Full-time · Posted Jan 8, 2026
- Experience
- 6+ years in software engineering, with 3+ years leading teams in SW/HW reliability, debug, diagnostic, failure analysis or related fields.
About the job
Lead the Kernel Reliability team in improving the reliability, diagnostics, and failure analysis of Cerebras compute clusters and production services.
What you'll do
- Set technical vision and roadmap for kernel-centric reliability across internal and customer-facing systems.
- Build and improve diagnostic and debug tools, and help operations teams reduce downtime through failure analysis.
- Lead and mentor engineers while collaborating with software, debug, ASIC, and hardware architecture teams on reliability improvements.
What they ask for
- 6+ years in software engineering, including 3+ years leading teams in software/hardware reliability, debugging, diagnostics, failure analysis, or related fields.
- Expertise in parallel and distributed programming, diagnostic/debug tools, and debugging distributed and parallel applications.
- Deep understanding of computer architectures.
- Strong background in monitoring and reliability engineering, including incident response and post-mortem analysis.
Tools
Kernels, GPU
AI focus
The role supports reliability of the company’s AI compute clusters and underlying inference and training services.
Model-written summary · the original posting is authoritative