.svg)
Machine Learning Engineer
This course builds upon and extends the DevOps methodology commonly used in software development to build, train, and deploy machine learning (ML) models. The course is based on the four-level MLOPs maturity framework. It focuses on the first three levels: initial, repeatable, and reliable. The course emphasizes the importance of data, models, and code for successful ML deployments. It demonstrates the use of tools, automation, processes, and teamwork in addressing the challenges associated with handoffs between data engineers, data scientists, software developers, and operations. The course also discusses the use of tools and processes to monitor and take action when model predictions in production deviate from agreed-upon key performance indicators.
Day 1
: 1. Introduction to MLOps
- Processes
- People
- Technology
- Security and governance
- MLOps maturity model
2. Getting Started with MLOps: Experimentation Environments in SageMaker Studio
- Bringing MLOps to experimentation
- Setting up the machine learning experimentation environment
- Demonstration: Creating and Updating a Lifecycle Configuration for SageMaker Studio
- Hands-On Lab: Provisioning a SageMaker Studio Environment Using the AWS Service Catalog
- Workbook: Getting Started with MLOps
3. Repeatable MLOps: Repositories
- Managing data for MLOps
- Version control for machine learning models
- Code repositories in ML
4. Repeatable MLOps: Orchestration
- ML pipelines
- Demonstration: Using SageMaker Pipelines to Orchestrate Model Building Pipelines
Day 2
4. Repeatable MLOps: Orchestration (continued)
- End-to-end orchestration with AWS Step Functions
- Hands-On Lab: Automating a Workflow with Step Functions
- End-to-end orchestration with SageMaker Projects
- Demonstration: Standardizing an End-to-End ML Pipeline with SageMaker Projects
- Using third-party tools to ensure repeatability
- Demonstration: Exploring Human-in-the-Loop During Inference
- Governance and security
- Demo: Exploring Security Best Practices for SageMaker
- Workbook: Repeatable MLOps
5. Reliable MLOps: Scaling and Testing
- Scaling and multi-account strategies
- Testing and traffic shifting
- Demo: Using the SageMaker Inference Recommender
- Hands-On Lab: Testing Model Variants
Day 3
5. Reliable MLOps: Scaling and Testing (continued)
- Hands-On Lab: Shifting Traffic
- Workbook: Multi-account Strategies
6. Reliable MLOps: Monitoring
- The Importance of Monitoring in Machine Learning
- Hands-On Lab: Monitoring a Model for Data Drift
- Operational considerations for model monitoring
- Addressing issues identified through the monitoring of machine learning solutions
- Workbook: Reliable MLOps
- Hands-On Lab: Building and Troubleshooting an ML Pipeline
- Explaining the benefits of MLOps
- Comparing and Contrasting DevOps and MLOps
- Evaluating the security and governance requirements for a machine learning use case and describing possible solutions and mitigation strategies
- Setting up experimentation environments for MLOps with Amazon SageMaker
- Explaining best practices for versioning and maintaining the integrity of ML model assets (data, models, and code)
- Describing three options for creating a complete CI/CD pipeline in a machine learning context
- Reviewing best practices for implementing automated packaging, testing, and deployment (data/model/code)
- Demonstrating how to monitor machine learning-based solutions
- Demonstrating how to automate a machine learning solution that tests, packages, and deploys a model automatically; detects performance degradation; and retrains the model using newly acquired data
This course includes instructor lecture, presentations, hands-on labs, demonstrations, and group exercises/discussions.
This course is intended for the following job roles:
- DevOps
- Machine Learning & AI
.svg)
