MLOps Engineer

Main Tasks

• Develop and maintain Inference Platform-as-a-Service
• Automate the lifecycle of ML models — from registration to model serving
• Create new products for automating ML development
• Develop platform services for ML
• Explore new platforms/tools for integration into company products
• Promote and develop best practices in MLOps

Expected (ML / Inference Track)

• Experience in deploying and operating Kubernetes for model inference (GPU/CPU)
• Practical experience with MLOps deployment and operation tools (Triton Inference Server, BentoML or similar)
• Knowledge of auto-scaling principles, load balancing, and traffic routing in the context of ML services
• Experience in configuring and using GPU infrastructure: drivers, CUDA Toolkit, MIG, GPU-Enabled Docker
• Proficient in Python

Expected (OPS Track)

• Expert knowledge of Kubernetes: controllers, operators, HPA, working with GPU
• Experience in infrastructure automation and Infrastructure as Code (Terraform, GitOps approaches)
• Experience with version control systems and CI/CD (GitLab, GitHub)
• Linux administration: from setup to monitoring and troubleshooting
• Experience in implementing and working with monitoring stack (Prometheus Stack)