Site Reliability Engineer - AI Infrastructure

A newly created opportunity exists for a Site Reliability Engineer to keep large‑scale GPU infrastructure fast, stable, and observable, working where reliability, networking, and platform engineering meet. The ideal candidate will build the automation, observability, and self‑service tooling that make live GPU infrastructure more reliable, while helping shape operational standards within a small, technically strong team. This is a fully remote role open to candidates anywhere in the UK.

• Own the reliability and performance of production GPU infrastructure, including the high‑performance networking that underpins it.
• Design, build and maintain observability, monitoring and dashboards. Improve signal quality through correlation, enrichment, suppression and deduplication.
• Build Python‑based automation for incident triage, runbook execution and routine operational tasks. Maintain version‑controlled runbook‑as‑code and automation libraries.
• Integrate observability, ITSM and infrastructure APIs to enrich alerts and automate end‑to‑end operational workflows.
• Deliver internal tools and self‑service capabilities, including CLI utilities, Chat Ops integrations and dashboards.
• Turn post‑incident learnings into better tooling and operational standards. Build safe, auditable automation for higher‑risk actions, with appropriate approval controls.

• Experience:
  
  A proven track record as an SRE or Production/Infrastructure Engineer, with hands‑on responsibility for GPU infrastructure in a live production environment.
• Core Tech/Domain:
  Production GPU infrastructure operations (for example, GPU cloud, AI/ML compute clusters or HPC). Strong Python for automation and tooling.
• Methodology/Protocols:
  Deep observability and dashboarding experience (for example, Prometheus, Grafana or similar), alert design and noise reduction, incident management, and runbook‑as‑code practices.
• Soft Skills:
  
  An ownership mindset, calm under incident pressure, and the discipline to fix root causes through automation rather than repeating manual fixes. You should be comfortable working autonomously in a remote team.

• Networking reliability experience, particularly low‑latency networking and Infini Band
• Platform engineering background (Kubernetes, infrastructure‑as‑code, internal developer platforms)
• HPC infrastructure exposure (for example, Slurm‑managed clusters or research computing environments)
• Chat Ops and ITSM integration experience (for example, Slack/Teams bots, Service Now or Jira Service Management APIs)

• Fully remote working (UK‑based)
• Full benefits package to be confirmed

• £120,000