Determine when to use and when not to use ML.
Rightsize resources (for example, instances, Provisioned IOPS, volumes).
Debug and troubleshoot ML models.
Transform data in transit (ETL, AWS Glue, Amazon EMR, AWS Batch).
Analyze and evaluate feature engineering concepts (for example, binning, tokenization, outliers, synthetic features, one-hot encoding, reducing dimensionality of data).
Detect and mitigate drops in performance.
Evaluate metrics (area under curve [AUC]-receiver operating characteristics [ROC], accuracy, precision, recall, Root Mean Square Error [RMSE], F1 score).
Perform cross validation.
ML on AWS (application services), Amazon Polly, Amazon Lex, Amazon Transcribe
Interpret descriptive statistics (for example, correlation, summary statistics, p-value).
Expose endpoints and interact with them.