AWS Machine Learning Specialty (MLA-C01) — Hands-On Prep
Stop memorizing flashcards. Build the systems the MLA-C01 exam asks about: S3 data lakes, Glue ETL pipelines, SageMaker training and deployment, Clarify bias detection, MLOps automation, and production monitoring. Every mission runs in a real AWS sandbox with automated validation.
Learning Paths
Data Depths
S3 data lakes, Glue DataBrew cleaning, Athena EDA, Glue ETL, and Kinesis real-time features. 5 missions for MLA-C01.
Vision Reef
Rekognition, Comprehend, Textract, Translate, Polly, and a document intelligence pipeline. 5 missions.
SageMaker Summit
SageMaker Studio, XGBoost training, real-time endpoints, batch inference, and Canvas no-code ML. 5 missions.
The Fairness Depths
Evaluation metrics, Clarify bias detection, SHAP explainability, and SageMaker Debugger. 4 missions.
Pipeline Forge
SageMaker Pipelines, Model Registry, CloudFormation deployment, serverless inference, and auto-retraining. 5 missions.
Sentinel Depths
Model Monitor, CloudWatch anomaly detection, VPC isolation, and cost optimization. 4 capstone missions.
All Labs
Build an S3 Data Lake with Versioning
Structure a production S3 data lake with versioning and lifecycle policies.
Clean Data with AWS Glue DataBrew
Build visual data cleaning pipelines with Glue DataBrew transforms.
SQL-Based EDA with Amazon Athena
Run exploratory data analysis on S3 data using Athena SQL queries.
Glue ETL Pipeline with Feature Engineering
Build ETL pipelines and engineer ML features with AWS Glue.
Real-Time Features with Kinesis Streaming
Stream and extract ML features in real-time with Amazon Kinesis.
Detect Objects in Images with Rekognition
Use Amazon Rekognition to detect objects, faces, and labels in images.
Text Analysis with Amazon Comprehend
Analyze sentiment, entities, and key phrases with Comprehend NLP.
Extract Documents with Amazon Textract
Pull text, tables, and forms from documents with Textract OCR.
Translate & Synthesize Speech with Polly
Build multilingual apps with Amazon Translate and Polly text-to-speech.
Build a Document Intelligence Pipeline
Chain Rekognition, Comprehend, and Textract into an AI document pipeline.
Set Up SageMaker Studio & Notebooks
Configure SageMaker Studio IDE and run your first ML notebook.
Train an XGBoost Model on SageMaker
Run a managed XGBoost training job with SageMaker built-in algorithms.
Deploy a Real-Time Inference Endpoint
Deploy your trained model to a SageMaker real-time endpoint.
Run Batch Inference with SageMaker
Process large datasets offline with SageMaker Batch Transform.
Build No-Code ML Models with Canvas
Create ML models without writing code using SageMaker Canvas.
ML Evaluation Metrics — Classification & Regression
Build confusion matrices and calculate precision, recall, F1, and RMSE.
Detect ML Bias with SageMaker Clarify
Run bias analysis on your ML model with SageMaker Clarify.
Explain ML Models with SHAP & Clarify
Generate SHAP feature importance explanations for ML predictions.
Debug ML Training with SageMaker Debugger
Detect overfitting, vanishing gradients, and training anomalies automatically.
Build a SageMaker ML Pipeline
Chain processing and training steps into an automated SageMaker Pipeline.
Model Registry with Versioning & Approval
Register, version, and approve ML models with SageMaker Model Registry.
Deploy Endpoints with CloudFormation
Provision SageMaker endpoints as infrastructure-as-code with CloudFormation.
Serverless ML Inference with Lambda
Deploy a scikit-learn model as a serverless Lambda API.
Automated ML Retraining Pipeline
Trigger model retraining automatically when new data lands in S3.
Monitor ML Models with SageMaker Model Monitor
Detect data drift and quality issues with Model Monitor data capture.
CloudWatch ML Dashboard with Anomaly Detection
Build ML observability dashboards with CloudWatch anomaly detection.
Secure SageMaker with VPC Isolation
Lock down SageMaker endpoints and training with VPC isolation.
Optimize ML Costs with Spot Training
Cut ML training costs with spot instances and endpoint auto-scaling.
Ready to start the MLA-C01 track?
Every lab runs in a safe AWS sandbox with automated validation. Complete labs show up in your verified portfolio.
Launch Cloud Edventures →