Responsibilities
- Implement and manage systems for deploying, monitoring, and retraining machine learning models.
- Build and maintain automated pipelines for deployment, inference, monitoring, and retraining workflows.
- Detect and respond to data and model drift using statistical and monitoring techniques.
- Track and log machine learning experiments to ensure reproducibility and auditability.
- Design and evolve end-to-end MLOps architecture aligned with business needs.
- Expose machine learning models as RESTful services for application integration.
- Research, evaluate, and integrate MLOps tools, platforms, and frameworks into data science workflows.
- Advance organizational maturity in MLOps by executing a defined backlog of improvement initiatives.
- Promote agile, automated practices within data science teams to improve efficiency and delivery speed.
- Deliver internal training sessions and presentations on MLOps tools and best practices.
Requirements
- Extensive hands-on experience with Kubernetes for container orchestration.
- Proven experience in operationalizing machine learning models using platforms such as Kubeflow, AWS SageMaker, Google AI Platform, Azure Machine Learning, DataRobot, or DKube.
- Solid understanding of machine learning and artificial intelligence concepts, including direct experience in developing ML models.
- Proficient in Python for both machine learning and automation scripting, with working knowledge of Bash and Unix command-line tools.
- Demonstrated experience in setting up and managing CI/CD/CT pipelines.
- Familiarity with cloud computing platforms, with a preference for AWS experience.
Responsibilities
- Model Deployment, Model Monitoring, Model Retraining
- Deployment pipeline, Inference pipeline, Monitoring pipeline, Retraining pipeline
- Drift Detection, Data Drift, Model Drift
- Experiment Tracking
- MLOps Architecture
- REST API publishing
Job Responsibilities
- Research and implement MLOps tools, frameworks and platforms for our Data Science projects.
- Work on a backlog of activities to raise MLOps maturity in the organization.
- Proactively introduce a modern, agile and automated approach to Data Science.
- Conduct internal training and presentations about MLOps tools’ benefits and usage.
