Professional Training · Applied AI Engineering Track
MLOps Engineering on AWS
A three-day course on operationalising machine learning, treating data, model and code as equally important to a successful deployment. It covers how MLOps differs from DevOps, end-to-end automation of ML workflows, the Amazon SageMaker features that support it, deployment strategies and inference options, and monitoring for data drift, bias, resource consumption and latency once models are live.
Authorised Instructors
Eric Tai
Head, Technology
- Machine Learning Engineer – Associate
- Security – Specialty
- DevOps Engineer – Professional
- Solutions Architect – Professional
- +4 more
Hui Liang
Chief Digital Officer
- Machine Learning Engineer – Associate
- Security – Specialty
- Generative AI Developer – Professional
- Solutions Architect – Professional
- +5 more
Any of these AWS Authorised Instructors may deliver this class, depending on the scheduled run. Trainers confirmed on enrolment.
Who Should Attend
- MLOps Engineers: Engineers automating the machine learning lifecycle.
- DevOps Engineers: Engineers extending delivery practice to ML workloads.
- Data Engineers: Engineers supplying and maintaining ML data pipelines.
- ML Platform Engineers: Staff building the platform models run on.
- Operations Staff: Teams responsible for models once they are in production.
Prerequisites
This is an intermediate course. To get the most from it, you should have:
Completion of AWS Technical Essentials, or an equivalent working knowledge of core AWS services.
Completion of DevOps Engineering on AWS, or equivalent CI/CD and automation experience.
Completion of Practical Data Science with Amazon SageMaker, or equivalent hands-on SageMaker experience.
Learning Objectives
In this course, you will learn to:
Define machine learning operations: Describe MLOps and how it differs from DevOps.
Map the ML workflow: Understand the stages and the handoffs between the teams involved.
Automate the pipeline: Build processes that build, train, test and deploy models automatically.
Use SageMaker for MLOps: Apply the key Amazon SageMaker features that support automation.
Choose a deployment approach: Weigh up inference options and deployment strategies for each use case.
Monitor live models: Detect data drift and bias and track resource consumption and latency.
Certification
A MY Code Campus certificate of completion is awarded for MLOps Engineering on AWS.
In-person in Kuala Lumpur or live virtual. Team & enterprise rates available.