MLOps Engineer

IT Factory is a dynamic company experienced in the development, implementation, and support of IT solutions for a significant Polish company in the insurance and finance industry.

• Minimum 3 years of experience in DevOps, MLOps, or software engineering, including production experience with ML models
• Advanced knowledge of Docker and Kubernetes (cluster management, Helm charts, Ingress)
• Deep knowledge of Azure (especially Azure ML, AKS, Azure Container Registry) or GCP/AWS with readiness to quickly adopt Azure
• Experience building CI/CD pipelines (Azure DevOps, GitHub Actions, Jenkins) with ML specifics (e.g., model training as a pipeline step)
• Good knowledge of Python (working with ML tool SDKs) and Bash/Shell scripting
• Practical experience with MLflow, Kubeflow, or native cloud solutions for model lifecycle management
• Knowledge of Terraform, Bicep, or Ansible
• Higher technical education (computer science, telecommunications, or related)
• "Automation First" approach aiming to eliminate manual work via scripts and tools
• Ability to work at the intersection of Data Science and IT Operations teams
• Proactivity in resolving performance issues and production incidents
• Service provision limited to the territory of Poland

Nice to have:

• Azure certifications: DevOps Engineer Expert (AZ-400) or Azure AI Engineer (AI-102)
• Experience deploying LLM models and RAG architectures
• Knowledge of monitoring tools (Prometheus, Grafana, Azure Monitor)
• Understanding of networking concepts in hybrid cloud (VPN, VNet, Private Endpoints)
• Knowledge of vector databases (e.g., in context of Azure AI Search)

• Creating and maintaining an AI platform in the insurance industry
• Designing and building MLOps/LLMOps infrastructure: scalable environment for training and serving models (including Azure Machine Learning, Azure AI Foundry, AKS)
• Implementing CI/CD/CT pipelines for ML solutions, including automatic testing, data and model versioning (DVC, MLflow), continuous training
• Preparing Docker images for AI/GenAI models and managing their deployments on Kubernetes clusters with hybrid architecture (integration with on-premise systems)
• Implementing model monitoring (detecting Data Drift/Model Drift), logging, and alerting to ensure high availability of AI services
• Providing technical support for compliance with AI Act: tools for model auditability, data lineage, and security (access management, encryption)
• Optimizing costs and performance: managing Azure resources, optimizing model inference time, scaling infrastructure based on load

Service provision limited to the territory of Poland.