MLOps Engineering on AWS
Field | Description |
Purpose | To bridge the gap between ML development and operational excellence by extending DevOps methodologies to the ML lifecycle, ensuring models are repeatable, scalable, and secure. |
Audience | Practitioners focused on the intersection of data science and operations who need to standardize and automate the path to production. |
Role | MLOps Engineers and DevOps Engineers. |
Domain | Machine Learning Operations (MLOps) / AI Engineering / CI/CD. |
Skill Level | Intermediate. |
Style | Experiential learning with a heavy focus on coding and architectural design, including SageMaker-based labs, maturity model workbooks, and automated retraining demonstrations. |
Duration | 3 Days. |
Related Technologies | Amazon SageMaker (Studio, Pipelines, Projects), AWS Step Functions, AWS Service Catalog, and Amazon CloudWatch. |
Course Description
MLOps Engineering on AWS builds upon the foundational DevOps methodology to specifically address the unique challenges of building, training, and deploying machine learning models. The course is structured around a four-level MLOps maturity framework, focusing on transitioning from manual processes to reliable, automated systems. You will learn how to unify the handoffs between data engineers, data scientists, and developers while maintaining the integrity of data, model, and code assets. The curriculum emphasizes real-world operations, including handling model drift and implementing automated retraining loops.
Who is this course for
This course is intended for:
MLOps Engineers who want to productionize and monitor ML models at scale within the AWS ecosystem.
DevOps Engineers responsible for the successful deployment, maintenance, and operational health of ML models in production.
Course Objectives
Strategy & Governance: Differentiate between DevOps and MLOps, and evaluate security/governance requirements for ML use cases.
Experimentation: Set up scalable experimentation environments using Amazon SageMaker Studio.
Integrity & Versioning: Apply best practices for versioning data, model, and code to ensure reproducible deployments.
CI/CD Orchestration: Describe and implement three distinct options for creating full CI/CD pipelines in an ML context.
Automation: Demonstrate how to automate the testing, packaging, and deployment of models.
Reliability: Detect performance degradation in production and trigger automated retraining on newly acquired data.
Prerequisites
Fundamental: AWS Technical Essentials (Classroom or Digital).
Advanced Operations: DevOps Engineering on AWS or equivalent professional experience.
ML Knowledge: Practical Data Science with Amazon SageMaker or equivalent experience in training models.