AWS Machine Learning Engineer
AWS Machine Learning Engineer
Duration in Hours : 40
Duration in Days: 20
AWS Machine Learning Engineer
Course Code: AWSMLE20CS
20 Hours of Theory
20 Hours of Lab
- Overview of ML concepts (supervised, unsupervised learning)
- AWS services for ML (SageMaker, EC2, S3, Lambda)
- Data cleaning and transformation using AWS Glue
- Feature extraction and selection
- Storing and managing datasets with Amazon S3 and AWS Glue
- Setting up Amazon SageMaker environment
- Data labeling with Amazon SageMaker Ground Truth
- SageMaker Notebooks
- Algorithms in Amazon SageMaker (XGBoost, linear regression, etc.)
- Training and tuning ML models with SageMaker
- Model hyperparameter tuning using SageMaker’s built-in features
- Building deep learning models with TensorFlow or PyTorch on AWS
- Training large-scale models on GPU instances (Amazon EC2)
- Amazon Elastic Inference and AWS Batch for scalable training
- Model evaluation metrics (accuracy, precision, recall, etc.)
- Cross-validation and split strategies
- Analyzing model performance with SageMaker Experiments
- Deploying models to production using Amazon SageMaker endpoints
- Real-time vs batch inference
- Automating deployment with AWS Lambda and API Gateway
- AWS CloudWatch for model monitoring
- Optimizing model performance and cost
- Model retraining and A/B testing
- Data security best practices on AWS
- AWS Identity and Access Management (IAM)
- Compliance requirements (GDPR, HIPAA)
- Building end-to-end ML pipelines using AWS Step Functions
- Automating workflows and scheduling tasks
- Designing, training, and deploying an ML model on AWS
- Evaluating and improving performance