LLM Ops Engineer - Scalable AI Infra & Reliability

Arada; Dubai; Vor Ort; AED 260.000 - 380.000; Vollzeit

Zusammenfassung

Arada in Dubai is seeking a senior DevOps/MLOps engineer to own the infrastructure, deployment pipelines and reliability of the AI/LLM stack across production features. You will manage model access, guardrails and cost governance while collaborating with AI engineers.

This role emphasizes IaC, containers, Kubernetes, and observability, with Azure as a cloud and a focus on safe, scalable AI services that power customer-facing apps and internal workflows.

Qualifikationen

• Strong background in DevOps/MLOps (CI/CD, IaC, containers, Kubernetes).
• Experience operating LLM/AI workloads in production, not just training pipelines.
• Familiarity with observability stacks (Grafana, Prometheus, Datadog, or similar).
• Scripting proficiency (Python, Bash) and Azure (or equivalent cloud) experience.
• Understanding of API gateways, rate limiting, and cost governance for AI services.
• Experience with Langfuse/LangSmith (nice to have).
• Security/compliance background for AI systems handling customer data (nice to have).
• Experience supporting multiple AI vendor integrations concurrently (nice to have).

Aufgaben

• Build and maintain CI/CD pipelines for AI application and agent deployment
• Stand up monitoring, logging, tracing, and alerting for LLM usage, latency, quality, and cost
• Manage model/API access, API keys, rate limits, and multi-vendor LLM integrations
• Implement guardrails, evaluation gates, and safety checks as part of deployment pipelines
• Own infrastructure-as-code for AI services (containers, orchestration, secrets management)
• Partner with AI Engineers to produce models, RAG pipelines, and agent workflows
• Drive cost optimization across inference, vector storage, and compute
• Own incident response and rollback procedures for AI-powered production features

Kenntnisse

DevOps

MLOps

CI/CD

IaC

Containers

Kubernetes

Observability

Python

Bash

Azure

Tools

Terraform

Docker

Kubernetes

GitHub Actions

Azure DevOps

Azure

Grafana

Prometheus

Datadog

Langfuse

LangSmith

PromptFoo

Azure API Management

Kong

Python

Bash

Key Vault

HashiCorp Vault

Jobbeschreibung

Arada in Dubai is seeking a senior DevOps/MLOps engineer to own the infrastructure, deployment pipelines and reliability of the AI/LLM stack across production features. You will manage model access, guardrails and cost governance while collaborating with AI engineers.

This role emphasizes IaC, containers, Kubernetes, and observability, with Azure as a cloud and a focus on safe, scalable AI services that power customer-facing apps and internal workflows.