AWS Sagemaker Data Science

Responsibilities:

• Assess existing machine learning models, workflows, and infrastructure ( Python( Anaconda) for migration to AWS SageMaker.
• Design and implement migration strategies for on-premises, other cloud platforms, or older SageMaker environments to target SageMaker services.
• Leverage various SageMaker services, such as SageMaker Studio, Pipelines, Model Registry, and Endpoints, to streamline the ML lifecycle and model deployment.
• Prepare and validate data for training and inference within SageMaker.
• Containerize models and dependencies using Docker and AWS ECR for efficient deployment on SageMaker.
• Develop and optimize inference scripts for various model types within SageMaker endpoints.
• Configure and deploy SageMaker endpoints for real-time and batch predictions, ensuring high availability and scalability.
• Implement MLOps best practices within SageMaker, including automated model deployment, monitoring, and versioning.
• Troubleshoot and debug issues during migration and post-migration phases.
• Collaborate with data scientists, software engineers, and other stakeholders to ensure successful migration and integration of models.
• Optimize resource utilization and costs related to SageMaker deployments.
• Stay updated with the latest SageMaker features and best practices.

Required skills and experience:

• Strong understanding of machine learning concepts and lifecycle.
• In-depth knowledge and hands-on experience with AWS SageMaker services, including Studio, Terraform Pipelines, Model Registry, Training, and Endpoints.
• Experience with Terraform/ Lambda and containerization for ML model deployment.
• Experience with migrating ML models from diverse environments to AWS SageMaker.
• Familiarity with AWS services like S3, ECR, Lambda, and IAM for supporting SageMaker workloads.