Site Reliability Engineer (AI Platform)

Location: Hybrid in Reading, England, United Kingdom, or Berlin, Germany

Company

hirify.global combines advanced technology with a global network of people to make unusable data usable and support AI-driven real-world impact.

What you will do

• Own the reliability of the AI Platform, including GPU-backed model-serving and inference infrastructure for ML and LLM workloads.
• Define and operate SLOs, on-call processes, and incident response for production models and services.
• Build observability for services and ML models, including drift monitoring, LLM tracing, evaluations, guardrails, metrics, and logging.
• Shape company-wide platform direction and create golden paths, reusable developer tooling, GitHub Actions, GitOps workflows, and Terraform modules.
• Package reusable infrastructure components such as Grafana, Istio, CloudNative tools, model registries, and feature stores.
• Embed security, compliance, cost governance, data-residency controls, PII protection, and audit trails into the platform.

Requirements

• 5+ years of experience in infrastructure engineering, DevOps, or SRE operating large-scale, highly available production systems with Kubernetes.
• Hands-on production experience with Kubernetes, Helm, Terraform or CloudFormation, and at least one major cloud provider; AWS is preferred.
• Good proficiency in Python, Go, or scripting for automation and tooling.
• Experience owning at least one infrastructure build end to end with measurable outcomes such as deployment time, MTTR, cost, adoption, or availability.
• Strong first-principles problem solving, cross-functional collaboration, and communication across global teams and time zones.
• Willingness to support 24x7 operational processes.

Nice to have

• Experience running ML workloads on Kubernetes, including GPU scheduling, capacity, and cost management.
• Production-scale model serving with KServe, Ray Serve, Triton, vLLM, or similar tools.
• Experience with MLOps tooling such as Kubeflow, MLflow, Feast, or Weights & Biases.
• Production LLMOps experience covering inference serving, prompt and version management, tracing, evaluations, drift, guardrails, and cost per request.
• Experience governing ML and LLM workloads with data-residency, PII, and audit requirements.

Culture & Benefits

• Mission-driven environment focused on economic and social impact.
• People-centric workplace supporting growth, well-being, belonging, and authentic self-expression.
• Global collaboration across diverse cultures and perspectives.
• Opportunities for professional development, continuous learning, and meaningful contribution.