MLOps Engineering on AWS

This course builds upon and extends DevOps methodology to enable the building, training, deployment, and monitoring of machine learning models on AWS. It focuses on applying MLOps principles to create reliable, repeatable, and scalable ML solutions in production environments.

The course is based on the four-level MLOps maturity framework and focuses on the first three levels: initial, repeatable, and reliable. It highlights the importance of data, model, and code in successful ML deployments and demonstrates how tools, automation, processes, and collaboration help address challenges across data engineering, data science, development, and operations. The course also covers monitoring techniques and actions to take when model performance in production drifts from agreed-upon key performance indicators.

Course objectives

In this course, you will learn to:

Explain the benefits of MLOps
Compare and contrast DevOps and MLOps
Evaluate security and governance requirements for ML use cases and describe mitigation strategies
Set up experimentation environments for MLOps using Amazon SageMaker
Explain best practices for versioning and maintaining the integrity of ML assets such as data, models, and code
Describe options for creating CI/CD pipelines in an ML context
Recall best practices for automated packaging, testing, and deployment of ML solutions
Demonstrate how to monitor ML-based solutions
Demonstrate how to automate ML solutions that test, package, deploy, monitor, and retrain models

Prerequisites

We recommend that attendees of this course have:

AWS Technical Essentials (classroom or digital)
DevOps Engineering on AWS, or equivalent experience
Practical Data Science with Amazon SageMaker, or equivalent experience

Target audience

This course is intended for:

MLOps engineers who want to productionize and monitor ML models on AWS
DevOps engineers responsible for deploying and maintaining ML models in production

Day 1 – Foundations and initial MLOps

The first day introduces MLOps concepts, including processes, people, technology, security, governance, and the MLOps maturity model. You work with experimentation environments in SageMaker Studio, including lifecycle configurations, provisioning environments, and managing data, models, and code repositories for repeatable MLOps.

Day 2 – Repeatable MLOps and orchestration

Day two focuses on orchestration and automation using services such as SageMaker Pipelines, AWS Step Functions, and SageMaker Projects. You explore governance, security best practices, human-in-the-loop patterns during inference, and techniques for scaling and testing ML solutions.

Day 3 – Reliable MLOps and monitoring

The final day covers reliable MLOps practices, including scaling strategies, traffic shifting, and testing model variants. You work with monitoring ML solutions for data drift, operational considerations for model monitoring, and remediation strategies, supported by hands-on labs and workbook activities.

Practical information

Duration: 3 days
Price: 27 900 NOK
Course level: Intermediate

FAQ

Er dette et sertifiseringskurs?
Nei, dette er et opplæringskurs og gir ingen formell sertifisering.

Er kurset praktisk rettet?
Ja, kurset inkluderer presentasjoner, hands-on labs, demonstrasjoner, kunnskapstester og workbook-aktiviteter.

Hvilke AWS-tjenester brukes i kurset?
Kurset bruker blant annet Amazon SageMaker Studio, SageMaker Pipelines, Step Functions og relaterte tjenester for MLOps.

Passer kurset for data scientists uten DevOps-erfaring?
Noe DevOps-erfaring anbefales, enten gjennom DevOps Engineering on AWS eller tilsvarende praktisk erfaring.

Dekker kurset hele MLOps-livssyklusen?
Kurset fokuserer på de første tre nivåene i MLOps-modenhetsmodellen og dekker eksperimentering, automatisering, drift og overvåking.

Price:	27,900 NOK
Included items:	Course documentation and exercises. Lunch and refreshments for in class events only.
Hours	09:00 - 17:00
Durance:	3 days
Languages	Norwegian, English