Director, AI Accelerator Infrastructure & SRE

d-Matrix; Santa Clara (CA); On-site; USD 250,000 - 350,000; Full time

Job summary

d-Matrix is seeking a Director of Site Reliability Engineering to architect and lead our AI accelerator infrastructure. You will own the end‑to‑end reliability of colocation, on‑premise lab clusters, and cloud environments (AWS/Azure/GCP), while shaping SRE as a discipline.

Lead 3–5 engineers, drive SLOs, RCA, and observability from day one. The role requires deep Linux, storage, multi‑cloud ops, and executive communication to align stakeholders with strategy and delivery.

Qualifications

• 15+ years in SRE, infrastructure engineering, or production engineering.
• Experience building or rebuilding an SRE function with measurable impact.
• Strong Linux, networking, and storage fundamentals at scale.
• Proven ability to define SLOs, on-call models, and observability from scratch.
• Executive communication to translate risk into clear narratives.

Responsibilities

• Build and lead SRE from the ground up, owning global infrastructure.
• Define charter, roadmap, and on-call model; hire 3–5 SRE engineers.
• Own end-to-end reliability across cloud, colocation, and on‑premises.
• Establish incident management, RCA, and continuous improvement.
• Own cost visibility, capacity planning, and FinOps across tiers.
• Drive IaC discipline and platform automation at scale.
• Partner with DevOps to align pipelines with reliability goals.

Skills

SRE leadership

Linux systems

IaC (Terraform, Ansible)

Kubernetes

Observability (Prometheus, Grafana, or

Education

Bachelor’s or Master’s in CS/EE or related field

Tools

Terraform

Ansible

Kubernetes

Prometheus

Grafana

Datadog

Job description

d-Matrix is seeking a Director of Site Reliability Engineering to architect and lead our AI accelerator infrastructure. You will own the end‑to‑end reliability of colocation, on‑premise lab clusters, and cloud environments (AWS/Azure/GCP), while shaping SRE as a discipline.

Lead 3–5 engineers, drive SLOs, RCA, and observability from day one. The role requires deep Linux, storage, multi‑cloud ops, and executive communication to align stakeholders with strategy and delivery.