AWS SageMaker

Job Summary: We are seeking a highly skilled SageMaker Unified Studio (SMUS) Migration & MLOps Engineer to lead the migration of existing AWS SageMaker AI environments into SageMaker Unified Studio while establishing enterprise-grade CI/CD, MLOps, and governance frameworks. The ideal candidate will possess deep expertise in AWS data and machine learning services, cloud migration strategies, MLOps automation, Infrastructure as Code (IaC), and secure multi-account AWS architectures. This role will partner with Data Scientists, Platform Engineers, Cloud Architects, Security teams, and DevOps teams to modernize ML development and deployment processes.

Key Responsibilities:

• SageMaker Unified Studio Migration
  • Lead migration of existing SageMaker AI environments into SageMaker Unified Studio (SMUS).
  • Design and implement SMUS project structures, including project profiles, blueprints, and workspace configurations.
  • Utilize AWS migration utilities, upgrade tools, and automation scripts to onboard existing workloads.
  • Assess, migrate, and modernize legacy notebook environments and ML workflows.
  • Integrate existing EMR on EC2 clusters and compute resources into SMUS.
• Data Platform Integration
  • Migrate and onboard Amazon S3 data assets into SMUS projects.
  • Configure and manage AWS Glue Data Catalog integration.
  • Implement data governance and access controls using AWS Lake Formation.
  • Develop automated onboarding processes for data sources, metadata, and catalogs.
• Workflow Orchestration & Scheduling
  • Design and implement notebook scheduling and workflow automation.
  • Configure Managed Workflows for Apache Airflow (MWAA) for orchestration.
  • Build parameterized, multi-notebook execution frameworks.
  • Replace legacy JupyterHub, Enterprise Gateway, and Livy-based architectures with modern SMUS-native capabilities.
• MLOps & CI/CD Engineering
  • Develop and maintain CI/CD pipelines using GitHub and GitHub Actions.
  • Implement reusable deployment workflows and automation templates.
  • Configure and manage SMUS CI/CD using:
    • aws-smus-cicd-cli
    • manifest.yaml
  • Enable automated testing, validation, deployment, and rollback capabilities.
• Cloud Security & IAM
  • Implement secure GitHub-to-AWS authentication using OIDC and JWT federation.
  • Design IAM roles, policies, and permission boundaries.
  • Define secure multi-account deployment models across Development, Testing, and Production environments.
  • Ensure compliance with enterprise security standards and governance requirements.
• Infrastructure as Code (IaC)
  • Develop AWS infrastructure using:
    • AWS CDK
    • AWS CloudFormation
  • Automate provisioning of networking, compute, data, and ML resources.
  • Create reusable infrastructure modules and deployment templates.
• Event-Driven Automation
  • Design event-driven deployment architectures.
  • Implement integrations leveraging:
    • Amazon EventBridge
    • AWS Lambda
    • AWS Step Functions
  • Automate ML workflow triggers and deployment promotions.
• Machine Learning Lifecycle Management
  • Build and maintain SageMaker Pipelines.
  • Implement Model Registry and approval workflows.
  • Configure MLflow for experiment tracking and model lineage.
  • Support model versioning, deployment, monitoring, and governance.
• Release Management & Governance
  • Manage Docker image lifecycle and Amazon ECR repositories.
  • Implement cross-account deployment and promotion strategies.
  • Integrate Veracode SAST scanning into CI/CD pipelines.
  • Establish approval gates and ServiceNow change-management traceability.
  • Maintain audit-ready deployment and compliance documentation.

Must-Have Skills: SageMaker Unified Studio (SMUS), AWS SageMaker / SageMaker AI, MLOps, AWS Glue, CI/CD, GitHub & GitHub Actions, Docker, AWS CDK / CloudFormation, Python, MLflow

Disclaimer: This job posting has been aggregated from external source. Role details, content, and availability are subject to change. Applicants are advised to confirm the latest information directly on the company website before applying.