Programmability for Snowflakes
Programmability is how code runs inside Snowflake. Stored procedures and user-defined functions can be written in SQL, JavaScript, Python or Java, and Snowpark provides a DataFrame API that pushes processing to the warehouse instead of pulling data out. Snowflake also hosts apps, including Streamlit, so logic and interfaces live beside the data.
Stage: Transformation · Part of the Snowflake roadmap · Last reviewed 19 November 2026
What to learn in Programmability
- Write a SQL user-defined function and call it in a query
- Create a stored procedure that runs several statements with control flow
- Build a Snowpark DataFrame transformation in Python
- Explain how Snowpark pushes work down to the warehouse
- Run a simple Streamlit app on top of a Snowflake table
Programmability interview questions
What is Snowpark?
Snowpark is a set of libraries and a runtime that lets developers write data processing in Python, Java or Scala using DataFrame-style APIs. The code is translated to SQL or runs in Snowflake's secure environment, so data stays inside the platform.
What is the difference between a UDF and a stored procedure?
A UDF returns a value and is called inside a query, typically per row or as a table function. A stored procedure is invoked with CALL, can run multiple statements, handle control flow and perform administrative or DML work, but is not used within a SELECT.
Where Programmability fits in the Snowflake roadmap
Programmability belongs to the Transformation stage (intermediate level). Build pipelines.
Other units in this stage
Back to the full Snowflake roadmap