AWS Certified Machine Learning Engineer – Associate

Overview of Data Ingestion and Storage

• 1. Know the Differences: AI vs Machine Learning vs Deep Learning
• 2. Types of data
• 3. Data warehouse vs Data Lakes vs Data Lakehouse vs Data Mesh
• 4. Cloud storage service on AWS
• 5. Hands-on: Creating and Configuring an S3 Lifecycle Policy
• 6. Hands-on: Instructions on how to use AWS Service Quotas

Amazon Glue, Amazon EMR, Amazon Kinesis Service, Amazon QuickSights and Amazon LakeFormation

• 1. Extract, Transform, Load (ETL) with Amazon Glue
• 2. Data Engineering with Amazon EMR
• 3. Amazon Kinesis Service
• 4. Amazon Lake Formation
• 5. Amazon QuickSights
• 6. Hands-on: Data Pipeline - AWS Glue for ETL, Integrating Athena for Querying the Processed Data
• 7. Hands-on: Build a Real Time Data Streaming System with Amazon Kinesis Data Streams
• 8. Hands-on: Preprocessing and normalizing data for ML with Glue Databrew

AWS Managed AI Services

• 1. Basic to Advance: 7 AWS AI Managed Services (Amazon Comprehend, Amazon Translate, Amazon Transcribe, Amazon Polly, Amazon Rekognition, Amazon Forecast, Amazon Lex + Connect)
• 2. Hands-on: Using AWS Transcribe to Batch Transcribe Hundreds of Audio Files
• 3. Hands-on: Building intelligent image analysis solutions on Customize Business

AWS Managed AI Services

• 1. Basic to Advance: 7 AWS AI Managed Services (Amazon Personalize, Amazon Textract, Amazon Kendra, Amazon Augmented AI, Amazon Lookout, Amazon Fraud Detector, Amazon Q Business/Developer)
• 2. Hands-on: Securing Your Code With Amazon Q Developer
• 3. Hands-on: Extract information from PDF/Image documents using Amazon Textract

Deep dive with Amazon SageMaker and Classification of Machine Learning Algorithms

• 1. Deep dive main feature on SageMaker Training & Deployment
• 2. Classification of Machine Learning Algorithms Amazon SageMaker's built-in machine learning algorithms (XGBoost, BlazingText, Object Detection, etc.)
• 3. Hands-on: A Step-by-Step Guide for Training a Basic Model using SageMaker
• 4. Hands-on: Sentiment Analysis of Text Files with Amazon Comprehend

Learn from real-world: Build high-level end-to-end Machine Learning project with SageMaker AI

• 1. Scenario/UseCase real-world
• 2. Introduction solution architecture
• 3. Explain flow
• 4. Implement solution in AWS follow the instructions.
• 5. Share estimate cost for architecture

Deep Learning, Large Language Models, Handling Unbalanced Data & Measuring Models

• 1. Foundational Deep Learning
• 2. Large Language Model
• 3. Handling Unbalanced Data & Measuring Model
• 4. Measuring Model
• 5. Hands-on: Deploy LLM DeepSeek in AWS
• 6. Hands-on: Build solution AI fairness and explainability with Amazon SageMaker Clarify

Deep dive Generative AI in AWS - Amazon Bedrock

• 1. AWS Foundation Models (Base Models)
• 2. Retrieval Augmented Generation (RAG)
• 3. Amazon Bedrock
• 4. LLM Agents
• 5. Hands-on: Using AWS Nova Multimodal Model to Generate Videos
• 6. Hands-on: Fine-Tuning and Deploying Custom AI Models on Amazon Bedrock

Learn from real-world 1: Build end-to-end solution RAG using Amazon Bedrock, implement AI Agent with Amazon Beckrock

• Learn from real-world 2: Building cost-effective RAG applications with Amazon Bedrock Knowledge Bases and Amazon S3 Vectors ( Amazon OpenSearch Service, Amazon Bedrock, Anthropic Claude...)
• 1. Scenario/UseCase real-world
• 2. Introduction solution architecture
• 3. Explain flow
• 4. Implement solution in AWS follow the instructions.
• 5. Share estimate cost for architecture

ML Implementation and Operations

• 1. Containers on AWS
• 2. AWS Service commonly used in MLOps
• 3. Overview Git/Git flow
• 4. Hands-on: Creating Responsible AI With Amazon Bedrock Guardrails
• 5. Hands-on: Implement best practice using AWS Temporary Credential with AssumeRole
• 6. Hands-on: Handle high dimensionality (>500-1000 feature) using SageMaker Data Wrangler

ML System Architecture

• 1. Well-Architecture machine learning lifecycle
• 2. MLOps Tools
• 3. Hands-on: Using FMs Stable Diffusion in AWS Bedrock to Generate Images
• 4. Hands-on: Spot Instances and Checkpointing and save up to 90% training cost

Security, Identity, and Compliance

• 1. Securing data in MLOps (AWS IAM, KMS, Macie, Secrets Manager, WAF, Shield,...)
• 2. Hands-on: Detecting PII in Amazon S3 with AWS Macie
• 3. Hands-on: Build GenAI Applications Using Amazon Bedrock With AWS PrivateLink To Protect Your Data Privacy

Management and Governance

• 1. AWS CloudWatch, AWS CloudTrail, AWS Config
• 2. AWS Budgets, AWS Cost Explorer
• 3. Cost Optimization Hub
• 4. Hands-on: Install CloudWatch agent using AWS Systems Manager
• 5. Hands-on: Utilizing Inference Components (ICs) to Host Multiple LLMs on SageMaker Real-Time Endpoints

AWS Certification Paths – Career counseling on MLOps role

• Learn from real-world: Low-Code/No-Code End-to-End Machine Learning for Startups - Amazon SageMaker Canvas
• 1. Scenario/UseCase real-world
• 2. Introduction solution architecture
• 3. Explain flow
• 4. Implement solution in AWS follow the instructions.
• 5. Share estimate cost for architecture

Practice Test 1 (65 câu)

Practice Test 2 (65 câu)