SRE DevOps Engineer (GenAI)

About the role

The SRE DevOps Engineer will serve as the operational backbone for TELUS Health's AI Automation team, ensuring that cutting‑edge GenAI solutions move from prototype to reliable production services. The role focuses on LLMOps, overseeing the performance, observability, and cost‑governance of large language model workloads.

Key responsibilities

• Maintain uptime and API latency for LLM inference services and vector database queries.
• Automate deployment of AI microservices (Python/NodeJS) and infrastructure using Terraform or similar IaC tools.
• Optimize cloud resource consumption on GCP/Azure for high‑compute AI workloads.
• Lead incident response, post‑mortem analysis, and implement self‑healing mechanisms for microservices.
• Ensure security, data privacy, and residency compliance for AI data stores and embeddings.
• Implement observability with tools such as Langfuse or Arize to monitor model responses, latency, and token usage.
• Monitor health of AI stack, including API rate limits, token consumption, and managed AI services availability.
• Execute model and software updates, handling minor change requests to keep solutions current.

Required profile

• 3+ years of experience in a DevOps role with exposure to software architecture and development.
• Strong understanding of large language models and the challenges of deploying GenAI at scale.
• Proven experience with GCP and/or Azure cloud platforms.
• Ability to diagnose root causes in complex, non‑deterministic AI systems beyond simple service restarts.
• Excellent communication skills to translate reliability metrics into actionable insights for AI developers and stakeholders.

Required skills

• GCP (including BigQuery, Looker) and Azure cloud services.
• Terraform or comparable Infrastructure‑as‑Code tools.
• Kubernetes (GKE/AKS) and Docker.
• Python and NodeJS programming.
• CI/CD pipeline implementation.
• Microservices architecture and API design.
• Vector databases and embedding pipelines.
• Observability platforms such as Langfuse or Arize.
• Vertex AI, OpenAI services, and related GenAI platforms.