AWS
AWS ML & cloud cert prep — the AWS MLA-C01 (Machine Learning Engineer Associate) learning path: four domain-weighted SageMaker decks and an 89-question scenario practice exam.
Flashcard decks
4 decks · 98 cards · in study order1AWS MLA · Data Preparation for ML (28%)The biggest domain: ingesting & storing data (S3, Kinesis, Glue), transforming & feature-engineering (Glue, Data Wrangler, DataBrew, EMR, Feature Store), labeling (Ground Truth), and ensuring data integrity, quality, and bias (Clarify) — plus the encoding, scaling, and splitting choices that decide whether a model can learn at all.302AWS MLA · ML Model Development (26%)Choosing an approach (built-in algorithms, JumpStart, Autopilot, or your own container), training & tuning (SageMaker training jobs, Automatic Model Tuning, distributed & spot training, the right instance types), and evaluating a model HONESTLY against the business goal — the metrics, overfitting controls, and error analysis the exam tests.263AWS MLA · Deployment & Orchestration (22%)Getting the model to production and keeping the pipeline automated: choosing the right SageMaker inference option (real-time, serverless, async, batch), sizing & auto-scaling compute, safe rollout strategies (blue/green, canary), and building repeatable ML CI/CD with SageMaker Pipelines, Model Registry, Projects, and IaC.214AWS MLA · Monitoring, Maintenance & Security (24%)Keeping a production model healthy and safe: monitoring models & data DRIFT (Model Monitor, Clarify), observability & auditing (CloudWatch, CloudTrail), cost/performance optimization, and SECURING the whole ML system (IAM least-privilege, VPC isolation, KMS encryption, Secrets Manager, and governance with Model Cards).21