MLOps Architect

Key Responsibilities:

• Design, build, and maintain scalable and reproducible ML pipelines using AWS SageMaker Pipelines.
• Automate the end-to-end ML lifecycle: data ingestion, model training, evaluation, deployment, and monitoring.
• Collaborate with data scientists to transition models from development to production.
• Manage model versioning, lineage, and experiment tracking using SageMaker Experiments or MLFlow.
• Implement monitoring and logging for deployed models using SageMaker Model Monitor and CloudWatch.
• Ensure CI/CD workflows for ML using CodePipeline, CodeBuild, or similar tools.
• Optimize infrastructure costs while maintaining model performance and uptime.
• Enforce ML governance, compliance, and security best practices.
• Contribute to documentation and promote MLOps best practices across teams.

Required Qualifications:

• 10 + years in an MLOps, DevOps, or Machine Learning Engineering role.
• Hands-on experience with AWS SageMaker (Training Jobs, Inference Endpoints, Pipelines, Model Registry).
• Proficient in Python, Boto3, Terraform or CloudFormation.
• Experience with containerization (Docker, ECS, EKS).
• Strong understanding of CI/CD pipelines for ML.
• Familiarity with ML frameworks such as TensorFlow, PyTorch, XGBoost.
• Experience in monitoring and logging tools like CloudWatch, Prometheus, or Grafana.

Preferred Qualifications:

• AWS certification (e.g., AWS Certified Machine Learning Specialty).
• Experience with feature stores (e.g., SageMaker Feature Store, Feast).
• Knowledge of data engineering tools such as AWS Glue, EMR, or Apache Airflow.
• Familiarity with governance and compliance standards for ML systems (e.g., model bias detection, GDPR compliance).