Snowflake Copilot: Exciting News — It’s Here!

Source: AI Generated Image

Finding it tough to remember SQL syntax?

Uncertain about your database’s data?

Wondering how to derive meaningful business insights and KPIs from your schema?

And if you’re new to Snowflake, writing optimized queries might seem like a challenge, right?

then answer would be enter Snowflake Copilot — a game-changer in the world of data management and analysis in snowflake. Today, I had the exciting opportunity to dive into Snowflake Copilot during its exclusive preview phase, available in just two regions. Let me tell you, It’s like having a helpful assistant for your data tasks.

Imagine a world where you no longer need to struggle with syntax or guess the contents of your database. With Snowflake Copilot, you can unlock the full potential of your data, effortlessly extracting valuable insights and KPIs to drive your business forward.

Whether you’re a Snowflake pro or a beginner, Snowflake Copilot makes data easier. It’s time to ditch the uncertainty and embrace better data management with Snowflake Copilot!

In this blog, I’ve shared my initial thoughts about Snowflake Copilot.

What is Snowflake copilot?

Well, it’s like having a supercharged assistant right within your Snowflake setup. This powerful tool, driven by LLM technology, is all about transforming how you work with databases. It makes analyzing data a breeze, helping you get the insights you need faster and with more depth.

Just picture it: You can ask questions directly to your Snowflake database and schema, and boom — get instant results. And here’s the kicker: Snowflake Copilot can even write SQL queries for you, including optimized ones if needed. It’s like having a data analysis wizard by your side, making the process smoother than ever.

In which regions is Snowflake Copilot available?

As of now, Snowflake Copilot is available in just two regions:

• AWS us-east-1
• AWS us-west-2

What are the costs associated with it?

As of now, there are no costs associated for interacting with Snowflake Copilot. However, if you run the queries generated by Snowflake Copilot within the engine, the warehouse costs apply as usual.

How to Set Up Snowflake Copilot?

After creating an account in the specified regions, when you select or create a session, Snowflake Copilot becomes available in the bottom right corner. Check out the screenshot below, where “ASK Copilot” is highlighted in the red box.

Fig 1 — Snowflake Copilot

Clicking on “Ask Copilot” gives you the option to select the database and schema you want to query. From there, you can inquire about tables or ask questions related to the database and schema.

Figure 2 — Copilot Workspace

Fig 3 — Database and Schema selection in CoPilot workspace

Here are the cutting-edge advantages I’ve experienced while using Snowflake Copilot:

1. Seamless dataset exploration: By asking plain English questions to Snowflake Copilot, we can effortlessly dive into our datasets.

For example: When prompted for a specific table, Snowflake Copilot intelligently discusses that table along with other relevant ones to derive valid KPIs.

Fig 4 — Dataset Exploration

2. Transforming Business Questions: With just one click, your business questions can be transformed into Snowflake queries and seamlessly embedded into the worksheet, as illustrated below,

Fig 5 — Transformation of Business Questions to SQL Queries with Steps

Here’s a step-by-step guide on using Snowflake Copilot:

Step 1: Choose the database and schema where you want to run your query.

Step 2: Phrase your question in plain English. For instance, you might ask, “Determine the number of orders placed by each customer, sorting them in descending order based on both the number of orders and the total amount spent by the customer.”

Step 3: Snowflake Copilot provides a detailed explanation based on your question and generates the corresponding SQL query.

Step 4: There are two buttons available:

• Add Button: Adds the SQL query to the worksheet without executing it against the database. Comments are also highlighted.
• Run Button: Adds the SQL query to the worksheet and executes it against the database. This step is highlighted in Step 5.

Step 5: Upon clicking the “Run” button, the SQL query is added to the worksheet and automatically executed against the database.

Step 6: The result set of your business question is displayed.

3. Validating Queries: If a query generated by Snowflake Copilot is invalid, it will be highlighted within the tool, as shown in the figure below.

Fig 6 — SQL Query Validation

Some Limitations Encountered with Snowflake Copilot:

• Cross-database and schema queries are not supported. To overcome this, alternative methods like creating views can be used for querying within Snowflake.
• Newly created databases and schemas may take some time to be reflected in Snowflake Copilot. If queried immediately after creation, you may encounter a message stating,

“We are in the process of indexing the tables and views in the selected database and schema. Please try again later.”

• Occasionally, Snowflake Copilot may take some time to generate a response.
• The SQL queries generated by Copilot may not always be 100% accurate, requiring tuning based on the output. However, overall functionality remains efficient.
• When listing tables for analysis, Copilot provides results for the top 10 tables and top 10 columns from each of those tables.

References: -

• https://www.snowflake.com/

About me:

I am a Cloud Data Architect with experience as a Senior Consultant at EY New Zealand. Throughout my career, I have worked on numerous projects involving legacy data warehouses, big data implementations, cloud platforms, and migrations. If you require assistance with certification, data solutions, or implementations, please feel free to connect with me on LinkedIn.