Snowflake Master Class 2026 | Core + SQL + API + Snowsight

Register for snowflake and start a free 30-day trial with 400 USD of free usage, choosing standard, enterprise, or business critical editions on Azure or Amazon Web Services.

Navigate the snow site interface for Snowflake, sign in with username and password, explore query building, data loading, virtual warehouses, databases, monitoring, data sharing, and account management with multi-factor authentication.

Explore the Snowflake classic console, retrieve records via worksheets from a fully qualified table, view run times and row counts, and export or preview schema details.

Understand how a Snowflake virtual warehouse supplies compute resources for query processing. Size determines credits per hour and performance, while auto suspend, auto resume, and per-second billing affect cost.

Understand how Snowflake multi-cluster warehouses scale compute to handle peak user and query load, using auto-scale and maximized modes, while managing credits and when to resize versus add more warehouses.

Learn to load data into Snowflake by connecting to an AWS S3 bucket, creating a stage, and building an external table from a CSV data source.

Explore Snowflake's cloud-native architecture, including storage, compute, and cloud services layers, and compare shared disk and shared nothing designs.

Explore Snowflake editions from standard to virtual private, detailing features like time travel, data encryption, disaster recovery, and HIPPA and Hitrust CSF compliance, plus edition switching and billing implications.

Revisit snowflake architecture and its three layers: data storage, compute, and cloud service, explain independence and the multi-cluster shared architecture, and emphasize cost optimization through efficient ETL and query tuning.

Explore Snowflake pricing: pay only for compute and storage you use, with decoupled resources, region-based credits, on-demand or capacity storage, and scalable warehouse sizing.

Switch to an account admin role to view usage data in the snow site and examine consumption, storage, data transfer, and credits across warehouses for billing.

Explore Snowflake’s built-in and custom roles, switch roles to see how views change, and learn how the six default roles and role hierarchy govern access.

Explore bulk loading and continuous loading with Snowpipe for Snowflake ingestion, using copy commands and virtual warehouses for batches, and stage-based pipelines with streams, tasks, and external tables for transformation.

Explore stages in Snowflake, comparing external stages for cloud storage loading via create stage with internal stages available by default, and learn about user and table stages and put command.

Learn to create a database, schema, an external stage, and an external table in Snowflake, using a public dataset with csv files and an MD5 key.

Import data into Snowflake and perform simple transformations with the copy command, including column selection, a computed is US flag, and country name extraction via substring.

Discover data transformation during import in Snowflake, including copying only selected columns, using autoincrement, and validating results with where clauses from S3 bucket and Excel sources.

Learn how to handle copy command errors in Snowflake production, using on_error options to continue, skip files, or abort, and review partial loads and error limits.

Explore creating and using file format objects in Snowflake, attaching them to stages, and overriding defaults in copy into commands for csv and json with skip header behavior.

Explore Snowflake copy options with validation_mode to validate data before loading, including return errors, return all errors, and enforce_length false, and learn how to test rows without copying.

Learn how to retrieve and analyze errors from a Snowflake copy into operation using validation mode and return underscore errors, and explore in-memory results via result_scan and error rows.

Explore the size_limit option for the copy into table command from external sources. The first file always loads; with multiple files, the limit prevents loading further data once exceeded.

Explore Snowflake copy command options underscore_length and truncate_columns, showing how to enforce length checks or automatically truncate long values during data load, with default behaviors and error handling.

Learn how copy into uses force to control reloading and duplicates, with force false by default, and how purge true removes files from the external stage after loading.

Learn to view load history for a table using the load history view and Copy_history, showing file, table, load time, status, and row counts for the last 14 days.

Snowflake handles semi-structured data as a first-class element, using a variant data type to load, store, and query JSON, Avro, XML, and more alongside structured data in SQL.

Create a Snowflake stage on S3, define a json external file format, and load json into a variant column, with partitioning on S3 by state.

Learn to retrieve data from hierarchical json arrays using snowflake's flatten function. Extract county and city from nested arrays, while handling quotes and json keys.

Learn to extract file names from json data and store them in a Snowflake table using union all, create or replace table, and insert into, verifying 1480 rows.

Load data from a parquet file by creating a stage with a url and parquet format, then query the staged files with a pattern to populate a raw table.

Explore how to load parquet files from an s3 stage, leveraging metadata like file name and row number, and capture timestamps across time zones to create a loading analytics table.

Snowflake uses automatic caching to speed repeated queries for up to 24 hours or until data changes, and suggests partitioning and avoiding select star, with same warehouse ETL for caching.

Practice Snowflake caching with a hands-on lab using the store_sales sample data, and learn how query profiles reveal table scans and performance gains through cached results.

Learn clustering in Snowflake to speed large data queries using parquet files and stages. Create and test a cluster by review date, compare caching and query profiler insights.

Explore how to evaluate and implement clustering in Snowflake using clustering keys, partition counts, and histograms, including testing keys with system$ clustering information and assessing reclustering costs.

Explore how Snowflake creates and alters cluster keys, then watches data reorder in the background to optimize partitions, depth, and micro partitions.

Learn how to create an Amazon S3 bucket, configure an IAM role and policy, and upload files, with region selection, ownership, privacy settings, and external ID security.

Learn to access a non-public dataset in Snowflake using a storage integration with S3, create a storage integration object, configure allowed locations, and create a stage to upload parquet data.

Update the AWS storage integration by adjusting the IAM role’s trust relationships, copying the ARN and external ID, and pasting the updated XML back into the AWS console.

Learn how Snowpipe automates data loading via cloud notifications from S3 using Amazon SQS, with four stages: create the stage, test the copy command, create a pipe, configure S3 notifications.

Configure an AWS S3 event notification for the Snowpipe public dataset bucket to trigger on all object create events for csv files, delivering notifications to the linked SQS queue.

Discover how to check a pipe's status with system$pipe_status, list pipes with show pipes, pause with pipe_execution_pause=true (notifications queue to S3 while paused), then resume to enable auto ingestion.

Learn how Snowflake time travel lets you view and restore previous table versions, using offset, timestamps, and query IDs to recover data after accidental updates.

Explore how to undo drops with the undrop time travel feature, restore dropped tables, schemas, and databases, and recover data using query ids and time-travel history.

Manage Snowflake retention time and time travel for tables by setting retention_time and understanding its impact on storage costs and data history.