MLOps Engineer (AI)

Location: Yerevan, Armenia

Company

hirify.global operates production machine learning and LLM workloads supported by GPU-based infrastructure.

What you will do

• Build and operate Kubernetes infrastructure for production GPU and machine learning workloads.
• Design scalable deployment patterns and reliable serving infrastructure for LLM inference and AI services.
• Own CI/CD pipelines, Helm charts, deployment automation, and infrastructure configuration.
• Design observability, metrics, SLOs, dashboards, and alerting with Prometheus, Grafana, and Alertmanager.
• Optimize GPU utilization, workload scheduling, autoscaling, batching, traffic routing, and container performance.
• Operate stateful services including Kafka, Redis, ClickHouse, and PostgreSQL; participate in on-call, incident response, root cause analysis, and corrective actions.

Requirements

• 3+ years of experience in MLOps, ML platform engineering, DevOps, or SRE, including 2+ years running GPU workloads in production.
• Strong Kubernetes experience with GPU device plugins, node selectors and taints, resource management, and HPA/KEDA autoscaling.
• Experience authoring and customizing Helm charts and owning CI/CD with GitHub Actions on self-hosted runners.
• Production observability experience with Prometheus, Grafana, and Alertmanager, including metrics, SLOs, and alert rules.
• Experience operating Kafka, Redis, ClickHouse, and PostgreSQL on Kubernetes, including backups and restore drills.
• Strong Docker, shell, and Python skills, plus production on-call and incident response experience.

Nice to have

• Bare-metal or on-premises GPU cluster setup, GPU Operator, driver lifecycle, NVLink, InfiniBand, local registries, and object storage.
• Cloud GPU infrastructure experience with Nebius, AWS, GCP, or Azure.
• Experience with Envoy Gateway, Gateway API, NVIDIA Dynamo, AIPerf, GenAI-Perf, or LLM-aware request routing.
• Knowledge of secrets management, RBAC, NetworkPolicy, image signing, scanning, and supply-chain security.
• Experience with MLflow, Kubeflow, Metaflow, Argo CD, Flux, MIG, or GPU time-slicing.

Culture & Benefits

• Collaboration with ML Engineering, Platform, and Infrastructure teams.
• Ownership of production infrastructure, reliability, scalability, and observability.
• Opportunity to evaluate technologies across GPU infrastructure, Kubernetes, LLM serving, and ML platform engineering.