Amazon Web Services
AWS Machine Learning Specialty
AWS Certified Machine Learning – Specialty (MLS-C01)
AWS's specialty certification for data scientists and ML engineers who build, train, tune, and deploy machine learning models on AWS. RETIRED: the last day to take MLS-C01 was March 31, 2026; AWS directs candidates to the Machine Learning Engineer – Associate (MLA-C01) instead. Existing certifications remain valid for 3 years from the date earned.
Blueprint-complete program: 4 decks, 887 cards.
What's on the exam
MLS-C01 Exam Guide (final version; exam retired March 31, 2026)
Data Engineering
20%Data repositories for ML (S3, lakes) · Data ingestion solutions (Kinesis, Glue) · Data transformation pipelines · Batch vs. streaming processing
Exploratory Data Analysis
24%Sanitizing and preparing data · Feature engineering · Handling missing data and outliers · Visualization and descriptive statistics · Labeling and ground truth
Modeling
36%Framing business problems as ML problems · Model selection (built-in algorithms, deep learning) · Training, tuning, and hyperparameter optimization · Evaluation metrics and validation · Overfitting, underfitting, and regularization
Machine Learning Implementation and Operations
20%Deploying and operationalizing models (SageMaker endpoints) · Performance, availability, and scaling · Security for ML solutions · Monitoring, retraining, and A/B testing · Choosing between AI services and custom models
Frequently asked questions
How much does the AWS Machine Learning Specialty cost?
The AWS Machine Learning Specialty costs $300. Final specialty-tier exam fee; exam retired March 31, 2026.
How long is the AWS Machine Learning Specialty and how many questions does it have?
65 items (50 scored, 15 unscored) — 180 minutes.
What do you need to pass the AWS Machine Learning Specialty?
Scaled score of 750 on a scale of 100–1,000.
Can you retake the AWS Machine Learning Specialty?
Not available — exam retired March 31, 2026; AWS recommends MLA-C01 (ML Engineer – Associate) instead.
What is the best way to study for the AWS Machine Learning Specialty?
Study the official blueprint, not random material: the exam is weighted by domain (Data Engineering 20%, Exploratory Data Analysis 24%, Modeling 36%, Machine Learning Implementation and Operations 20%). Spaced-repetition flashcards built domain-by-domain against that blueprint are the most time-efficient way to cover everything the exam tests.
Ready to prepare?
Our AWS Machine Learning Specialty program follows this exact blueprint — every domain covered, scheduled by spaced repetition, with a journey from first card to test day.