Snowflake SnowPro Advanced: Data Scientist (DSA-C03) Lessons
16 lessons, one per exam-guide subdomain, in the order the guide teaches them. Every claim is cited to the official documentation.
Domain 1: Data Science Concepts
4 lessons · 17% of the exam
1.1Supervised, Unsupervised and Reinforcement Learning in Snowflake ML
Subdomain 1.1: Define machine learning concepts for data science workloads.
15 min read
1.2Machine Learning Problem Types: Regression, Classification, Forecasting, Clustering
Subdomain 1.2: Identify machine learning problem types.
16 min read
1.3ML Lifecycle in Snowflake: Data Collection, Exploration, Feature Engineering and Training
Subdomain 1.3: Summarize the machine learning lifecycle.
2 pages · 22 min read
1.4Statistical Concepts: Distributions, CLT, Z/T Tests, Bootstrap, CIs
Subdomain 1.4: Define statistical concepts for data science.
16 min read
Domain 2: Data Preparation and Feature Engineering
4 lessons · 27% of the exam
2.1Snowpark DataFrames for Data Cleaning: Lazy Evaluation, Selecting Fields and Casting Types
Subdomain 2.1: Prepare and clean data in Snowflake.
2 pages · 21 min read
2.2Data Profiling, Notebooks and Descriptive Statistics in Snowflake
Subdomain 2.2: Perform exploratory data analysis in Snowflake.
2 pages · 22 min read
2.3Feature Engineering with Snowpark DataFrames and Snowflake ML Preprocessing
Subdomain 2.3: Perform feature engineering on Snowflake data.
2 pages · 19 min read
2.4Statistical Summaries and Outliers in Snowsight with SQL
Subdomain 2.4: Visualize and interpret the data to present a business case.
2 pages · 20 min read
Domain 3: Model Development
5 lessons · 31% of the exam
3.1Snowpark Python Sessions, Languages and Snowpark ML Connections
Subdomain 3.1: Connect data science tools directly to data in Snowflake.
2 pages · 19 min read
3.2Cortex AI Functions: Task-Specific Models and Prompt Engineering
Subdomain 3.2: Leverage GenAI and LLM models in Snowflake.
2 pages · 21 min read
3.3Snowflake Data Science Pipelines: Dynamic Tables, Python UDFs, UDTFs and Stored Procedures
Subdomain 3.3: Train a data science model.
2 pages · 25 min read
3.4Validating Models in Snowflake: Confusion Matrix, Thresholds, ROC and Regression Metrics
Subdomain 3.4: Validate a data science model.
16 min read
3.5Model Interpretation in Snowflake: SHAP, Feature Importance, Dependence and Intervals
Subdomain 3.5: Interpret a model.
16 min read
Domain 4: Model Deployment
3 lessons · 25% of the exam
4.1Snowflake Model Registry: Log, Promote and Serve Models
Subdomain 4.1: Move a data science model into production.
2 pages · 23 min read
4.2Model Monitoring in Snowflake: Drift, AUC, Precision, Recall and RMSE
Subdomain 4.2: Determine the effectiveness of a model and retrain if necessary.
17 min read
4.3Snowflake Model Registry: Versions, Metrics and Metadata Tags
Subdomain 4.3: Outline model lifecycle and validation tools.
2 pages · 19 min read