Job Details
Skills And Tools:
Job Description
The MLOps Engineer is responsible for automating, operationalizing and managing the machine learning lifecycle across all phases—training, evaluation, deployment, and monitoring. The role includes building CI/CD pipelines for ML workloads, enabling continuous training and deployment via Azure DevOps, maintaining feature and model registries and enforcing ML governance.
Overall Responsibilities:
• Design and implement end-to-end MLOps pipelines for ML model lifecycle management.
• Collaborate with Data Scientists to streamline model experimentation and deployment workflows.
• Ensure reproducibility, scalability, and automation of ML systems.
• Maintain production-grade infrastructure with focus on availability, monitoring, and fault-tolerance.
• Establish model governance mechanisms including audit trails, access controls, and compliance frameworks.
• Enable secure and ethical AI practices aligned with FATE (Fairness, Accountability, Transparency, Ethics).
• Contribute to improving code quality, process automation, and DevOps culture in AI teams.
Technical Responsibilities:
• Develop and maintain CI/CD pipelines using Azure DevOps, Git, and Azure Pipelines.
• Implement model training, evaluation, and deployment workflows using MLFlow, DVC, and Airflow.
• Manage model versioning and experiment tracking, enabling reproducibility and lineage.
• Automate testing using frameworks like pytest, behave, and integrate SonarQube for code quality.
• Design and maintain deployment strategies: Blue-Green, Canary, and Shadow deployments.
• Configure monitoring and alerting pipelines using Prometheus, Grafana, and email triggers.
• Enable feedback loops and retraining mechanisms triggered by concept or data drift.
• Ensure rollback and recovery strategies for deployed models.
Tools & Technologies:
• Version Control & CI/CD: Git, Azure DevOps, Azure Pipelines, DVC
• Experiment Tracking & Registry: MLFlow, DVC, Azure ML
• Testing: pytest, behave, SonarQube
• Orchestration: Airflow, Azure Data Factory (optional)
• Monitoring & Alerting: Prometheus, Grafana, Cloudera tools, email notifications
• Deployment: Docker, Kubernetes (optional), Azure ML Endpoints
• Programming: Python, Bash, YAML, JSON
• Storage & Compute: Azure Blob, Cloudera, HDFS
Job Requirements
Preferred Experience:
• 7-8+ years of hands-on experience in MLOps, DevOps, or ML Engineering roles.
• Proven experience deploying ML models at scale in production environments.
• Familiarity with monitoring model performance and automating drift detection and retraining workflows.
• Understanding of responsible AI concepts like fairness, transparency, and auditability.
Education & Certifications:
• Bachelor’s or Master’s degree in Computer Science, Data Engineering, or related field.
Certifications preferred:
Azure DevOps Engineer Expert
Certified MLOps Professional (TWiML, Coursera, or similar)
Azure AI Engineer Associate (Optional)
