MLOps Engineer
An MLOps Engineer operates where Machine Learning, Software Engineering, and DevOps intersect. Their role is to merge best practices from each discipline to deploy, scale, and maintain ML models efficiently in real-world production settings.
Key Responsibilities:
- Leadership in MLOps Strategy: Provide guidance on MLOps design principles, recommend tools and workflows, and lead their implementation within a large-scale enterprise environment.
- Managing Technical Debt: Identify and resolve existing inefficiencies and legacy issues in ML systems, while embedding MLOps best practices moving forward.
- Model Operationalization: Oversee the deployment of ML models, enhancing deployment workflows through automation.
- Automation & Quality Assurance: Design and maintain automated pipelines; integrate unit tests, validation steps, and CI/CD practices to ensure smooth model release cycles.
- Monitoring and Lifecycle Management: Track model performance and data integrity in live systems; collaborate with Data Scientists to retrain models and maintain retraining processes.
- Security and Governance: Ensure compliance with security protocols and cloud best practices for data and model management.
- Cross-Functional Collaboration:
- Partner with Data Engineers and modelers to align with data architecture.
- Work alongside Data Scientists to transition experimental models into production-ready systems.
- Coordinate with Software Engineers and IT teams to implement and support ML infrastructure.
Required Experience and Skills:
- Experience: 7-10 years in software engineering, data engineering, or MLOps, ideally in complex, enterprise environments.
- End-to-End MLOps Expertise: Hands-on experience designing and building MLOps frameworks from the ground up.
- Cloud Proficiency: Deep familiarity with AWS services (e.g., Athena, Glue, ECS, EKS, VPC) and deploying models using SageMaker.
- CI/CD Implementation: Experience automating deployments via tools like Azure Pipelines.
- Workflow Orchestration: Proficient with scheduling and monitoring tools such as Apache Airflow.
- Programming & Scripting: Strong coding abilities in Python and Bash; skilled in SQL and PySpark for data preparation.
- Infrastructure-as-Code: Working knowledge of Terraform or AWS CloudFormation for provisioning infrastructure.
- Security & Compliance: Understanding of cloud security standards and data protection measures.
- ML Framework Familiarity: Comfortable with tools like scikit-learn, PyTorch, and TensorFlow, and knowledgeable about the machine learning lifecycle.
Bonus Skills:
- Experience with Azure Synapse and Azure ML Studio.
- Hands-on familiarity with Databricks.
- Exposure to dbt (data build tool).
For a confidential discussion, please reach out to me at fiona.wong@sanderson-ikas.sg
Personal data collected will be used for recruitment purposes only.
Only shortlisted candidates will be notified / contacted.
EA Registration No: R21101138
"Sanderson-iKas" is the brand name for iKas International (Asia) Pte Ltd, a company incorporated in Singapore under Company UEN No.: 200914065E with EA license number 16S8086.
Website: www.sanderson-ikas.sg