Ways of Working with Snowflake
1. Ways of Working with Snowflake
Snowflake doesn't have only one way in, it has three. Knowing which one to reach for depends on what you're trying to do. In this video, you'll meet all three through a scenario that will feel familiar once you're working with real data.2. Three interfaces, three different jobs
The three primary ways to work with Snowflake are Snowsight in the browser, VS Code with the Snowflake extension in your editor, and the Snowflake CLI on the command line. We've summarized each interface at a high level, but we'll dive deeper later in this video. Now, it's Monday morning at Snowy Peak. A dashboard is showing numbers from last Friday. Something didn't run, or a query hit an error, or the warehouse wasn't available when it should have been. You need to get into Snowflake and figure out what has happened. And the right tool for this situation is Snowsight.3. Snowsight
Snowsight is the browser UI, it's where most people spend the majority of their time. When you log in, you should see the Home screen. The interface is split into four key sections: Section 1 is the navigation which is organized into three sections: Work with data, Horizon Catalog and Manage. Section 2 acts as a universal search bar; section 3 includes some quick actions to get the user working with their data straight away. Finally, section 4 includes all your recently viewed project and work for easy access. Work with data is where you'll spend the majority of your time: Projects, Ingestion, Transformation, AI and ML, Monitoring and Marketplace.4. Snowsight: Workspaces
When you navigate to Project, you'll have several options - the most commonly used is Workspaces; which is the default development environment for Snowflake today. A workspace is a private, file-based editor where you can create SQL, Python and Notebook files, organize them into folders, and work across multiple files side by side. If you're on an older account, you may see Worksheets instead — that's the legacy editor. Both work, but Workspaces is where Snowflake is heading.5. Snowsight: Context
At the top of any open file, you'll see your context dropdowns: role, warehouse, database, and schema. The database and schema work together, when you pick a database you also choose a schema (or the default for that database). Set these before you run anything. Wrong context is the number one reason a query fails when the SQL looks perfectly fine. By default the role and warehouse may be pre-selected, but ensure you're setting your database and schema within the context dropdowns or via SQL.6. Snowsight: Query History
Under Monitoring, Query History shows you everything you've run against your account: its status, how long it ran, who triggered it, and which warehouse it hit. If a query is missing from the list, open Filters in Query History. Options like Don't consume query credits can hide some runs, so adjust the filters and click Apply to see the full picture.7. Query History: Simplified Example
If you were investigating a failed pipeline, this would be your first stop. This is how you figure out exactly what broke and when. This is one of those features that feels like background noise until something goes wrong, then it becomes one of the most useful pages in Snowflake.8. VS Code with the Snowflake extension
Now let’s try a different scenario, Snowy Peak's data engineer is building a dbt model and needs to verify the SQL against live data. They could open a browser, paste the query into a worksheet, switch back to their editor, but that’s not the most efficient way. Instead they should run it in VS Code with the Snowflake extension. First, install the Snowflake extension from the marketplace, authenticate once, and you can execute SQL directly from inside your project.9. VS Code Extension: Simplified Example
No need to switch tools, the query lives next to the code it belongs with. This is the right tool when you're in a development workflow: writing, testing, and iterating all in one place.10. Snowflake CLI
Now for the final scenario. Snowy Peaks' data pipeline runs a validation script before every deployment: checking row counts, nulls, date ranges. No humans are clicking run. This needs to happen automatically.11. CLI: Simplified Example
That's what the CLI is for. The Snowflake CLI replaces the older SnowSQL. It lets you manage Snowflake objects, deploys apps, run queries and interact with Snowflake services directly from your terminal. It fits inside CI pipelines, scheduled jobs, and really anything that runs without a person in the loop.12. Let's practice!
We'll be focusing on the Snowsight interface throughout this course. Let's give it a try!Create Your Free Account
or
By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.