MLOps Engineer

We are looking for a hands-on MLOps Engineer with strong experience in deploying, scaling, and managing AI workloads in production environments. The role focuses on building robust AI infrastructure and enabling enterprise-scale adoption of Generative AI solutions.

Responsibilities:

• Deploy and manage open-source LLMs (Llama, Mistral, Qwen, etc. ) on Azure Kubernetes Service (AKS).
• Build and maintain model serving and routing infrastructure.
• Design scalable, reliable, and highly available AI endpoints.
• Implement model evaluation frameworks and performance monitoring.
• Drive observability, logging, and operational excellence for AI workloads.
• Manage usage monitoring, quota management, and cost optimisation.
• Collaborate with AI/ML teams to productionize GenAI solutions.
• Automate deployments through CI/CD pipelines.
• MLOPS - Should have strong exposure in hosting open-source models on Azure Kubernetes cluster, building model routers, OSS model evaluation, endpoint availability, usage monitoring, quotas and cost controls.

Requirements:

• Strong experience with Azure Cloud and Azure Kubernetes Service (AKS).
• Hands-on expertise in Kubernetes, Docker, and Helm.
• Experience hosting and serving Open Source LLMs.
• Exposure to model serving frameworks such as vLLM, Triton, KServe, TGI, or Ray Serve.
• Strong Python development and automation skills.
• Experience with monitoring and observability tools such as Prometheus, Grafana, and Azure Monitor.
• Understanding of GPU infrastructure, scaling, and cost optimisation.