Principal SRE: AI Inference Reliability & Scale

Cerebras; Sunnyvale (CA); On-site; USD 260,000 - 380,000; Full time

Job summary

Cerebras Systems is seeking a Principal SRE to define and drive the architecture for scaling our AI inference fleet across datacenters and cloud-based solutions. You will build self-service platforms, observability, and automation, enabling product teams and customers to operate with strong guardrails.

In the first year, you will lead a transformation from ops-centric reliability to a shared engineering discipline, mentor senior engineers, and shape capacity management and rollout safety.

Qualifications

• 15+ years in SRE or related infra, with large-scale reliability improvements.
• Experience building production control planes, schedulers, and orchestration.
• Ability to lead complex programs and mentor senior engineers.
• Proven cross-functional collaboration with product and customers.
• Hands-on with observability, incident response, and postmortems.

Responsibilities

• Define and deliver scalable, reliable infra across multi-datacenter and cloud.
• Architect self-service platforms with safe workflows.
• Establish SLOs/SLIs, error budgets, and capacity forecasting.
• Mentor senior SREs and drive automation of toil.
• Track impact with deployment velocity and MTTR metrics.

Skills

SRE

Infrastructure engineering

Platform engineering

Observability

Incident response

SLOs/SLIs

Cross-team leadership

Tools

Bazel

Job description

Cerebras Systems is seeking a Principal SRE to define and drive the architecture for scaling our AI inference fleet across datacenters and cloud-based solutions. You will build self-service platforms, observability, and automation, enabling product teams and customers to operate with strong guardrails.

In the first year, you will lead a transformation from ops-centric reliability to a shared engineering discipline, mentor senior engineers, and shape capacity management and rollout safety.