Software Engineer, Kernel Reliability

Job Title

Software Engineer, Kernel Reliability

Role Summary

Hands-on software engineer on the Kernel Reliability team focused on improving reliability of Cerebras compute clusters and associated inference, training, and production services. The role works close to code and hardware teams to design tooling and processes that reduce downtime and speed failure analysis.

Experience Level

Mid-level. Candidates with strong systems fundamentals are expected; new college graduates are welcome.

Responsibilities

Deliver kernel-centric reliability improvements, debugging tools, and operational support for on-field systems.

• Define and execute technical roadmap for kernel and system reliability.
• Partner with System and Cluster Operations to reduce downtime via tooling, analysis, and hands-on debugging.
• Enhance debug tools and instrumentation to accelerate failure analysis.
• Collaborate with software teams to harden the software stack and improve on-field observability.
• Work with ASIC and hardware architecture teams to design for reliability and debuggability.
• Participate in incident response, root-cause analysis, post-mortems, and drive measurable reliability improvements.

Requirements

Core qualifications and technical skills required to perform the role.

• Must-have: Strong programming in C/C++ and Python.
• Must-have: Solid foundations in operating systems, computer architecture, and systems programming fundamentals.
• Must-have: Ability to debug complex issues using logs, traces, and standard debugging workflows; interest in root-cause analysis.
• Nice-to-have: Exposure to parallel and distributed programming (message passing, multicore, GPU, embedded).
• Nice-to-have: Experience with debug and diagnostic tools (debuggers, core dump handling, tracing, sanitizers, profilers).
• Nice-to-have: Familiarity with debugging distributed/parallel applications (deadlocks, race conditions) and monitoring/incident response practices.

Education Requirements

Not specified.

About the Company

Company: Cerebras

Headquarters: Sunnyvale, CA, USA

Developer of wafer-scale AI accelerators, Cerebras designs the Wafer Scale Engine (WSE)-one of the world’s largest AI chips-to deliver high-speed training and inference solutions for model labs, enterprises, and AI-native startups.