Snowflake Tools

Quick answer

Snowflake tools are the interfaces, clients, libraries and partner products you use to load, transform, query, secure and report on data in Snowflake, from Snowflake's own Snowsight, Snowflake CLI and Snowpark to third-party tools such as dbt, Airflow, Fivetran, Informatica, Tableau and Power BI. Snowflake itself is a cloud data platform, not a single tool. Around it sits a toolchain with clear layers: ingestion, transformation, orchestration, BI, DevOps and monitoring. In the 14 Snowflake job descriptions we read on 25 October 2026, SQL appeared in all of them, while dbt and Python each appeared in 9.

Key takeaways

1. Snowflake tools at a glance

Short answer: Snowflake tools fall into eight layers, from the interface you type SQL into, through loading, transforming and scheduling, to reporting, deployment and cost control.

If you are still asking "what is Snowflake?", read our explainer on what Snowflake does first. In short, Snowflake is a cloud data platform with separate storage, compute (virtual warehouses) and cloud services layers. A virtual warehouse is a cluster of compute you start, stop and resize, and it uses credits, Snowflake's unit of compute billing. The tools below are how you actually work with that platform.

Two quick definitions you will see again. ELT means extract, load, then transform inside the warehouse, which is the usual pattern on Snowflake. ETL transforms data before loading it. Most modern Snowflake teams load raw data first and transform it with SQL or dbt.

2. Snowflake's own interfaces: Snowsight, Snowflake CLI and SnowSQL

Short answer: Snowsight is the web interface, Snowflake CLI is the preferred command-line client, and SnowSQL is the older command-line client that still works but gets no new features.

Snowsight

Snowsight is Snowflake's browser-based interface. You write SQL in worksheets, browse databases, create warehouses, load small files through the stage upload screen, open Notebooks and build Streamlit apps. For a fresher, Snowsight is where every first query happens. Our Snowflake account setup guide walks through it step by step.

A few Snowsight facts trip people up (Snowflake documentation, October 2026):

Snowflake CLI

Snowflake CLI is an open-source command-line client (the command is snow). It runs SQL and manages Snowpark functions and procedures, Streamlit apps, warehouses and Snowflake Native Apps (Snowflake documentation, October 2026). It is the tool you use in scripts and CI/CD pipelines, for example snow sql to run a query, or snow dbt execute to run a dbt project inside Snowflake.

SnowSQL

SnowSQL is the original command-line client. Snowflake's SnowSQL page calls it a legacy client in maintenance-only mode: new features go only to Snowflake CLI, which is the preferred client for new installs. Support ends by minor version, and the last line, SnowSQL 1.5.x, is supported until 16 April 2028 (Snowflake documentation, October 2026). So SnowSQL is not deprecated, and you will still meet it in older projects and some job descriptions, but start new work with Snowflake CLI.

Notebooks, Workspaces and the VS Code extension

Snowflake Notebooks are cell-based notebooks inside Snowsight where you mix Python, SQL and Markdown. They run on a Warehouse Runtime (GA) or a Container Runtime on Snowpark Container Services compute pools with CPU or GPU images. The original notebooks are now called "Legacy Notebooks", and Snowflake describes Jupyter-compatible "Notebooks in Workspaces" as the next generation (Snowflake documentation, October 2026).

Snowflake also publishes a Visual Studio Code extension with its own documentation page. We have not verified its feature list, so check the docs page before relying on a specific feature. VS Code itself does show up in hiring: one Hyderabad administrator job description we read in October 2026 asked for hands-on experience with an AI coding assistant and VS Code.

3. Developer tools: drivers, connectors, Snowpark and Streamlit

Short answer: Drivers connect applications and BI tools to Snowflake, Snowpark lets you write Python, Java or Scala that runs inside Snowflake, and Streamlit in Snowflake turns Python scripts into shared data apps.

Drivers and connectors

Every outside tool that talks to Snowflake, from Power BI to Airflow to your own Python script, goes through a driver or connector. Snowflake's drivers page lists official drivers for Go, JDBC, .NET, Node.js, ODBC (version 4.x is GA), PHP PDO and the Snowflake Connector for Python (GA). A shared Rust library called "Universal Core" now sits underneath ODBC 4.x and the Python connector (Snowflake documentation, October 2026).

In practice:

Snowpark

Snowpark is a set of libraries for Python, Java and Scala. You write DataFrame code on your laptop or in a notebook, and Snowpark turns it into SQL that runs on Snowflake compute. The operations are lazy: nothing runs until an action such as collect is called, and then the work is pushed down to a warehouse (Snowflake documentation, October 2026). This matters because your data never leaves Snowflake.

Snowpark also covers pandas on Snowflake (pandas code on Snowflake data, built on Modin), Snowpark Container Services (a managed container platform inside Snowflake, GA in AWS, Azure and Google Cloud commercial regions with some exceptions) and Snowpark-optimized warehouses for memory-heavy work. Our full guide to Snowpark in Snowflake covers setup, DataFrames, UDFs and use cases.

Here is a simplified example of opening a Snowpark Python session, based on the connection pattern in Snowflake's Snowpark documentation (October 2026). Once you have a session, you build DataFrames on it, and the SQL runs only when you call an action such as collect.

# simplified example: open a Snowpark Python session
from snowflake.snowpark import Session

connection_parameters = {
    "account": "<your snowflake account>",
    "user": "<your snowflake user>",
    # add your authentication settings here (for example key pair or SSO)
    "warehouse": "<your snowflake warehouse>", # optional
    "database": "<your snowflake database>", # optional
    "schema": "<your snowflake schema>", # optional
}
new_session = Session.builder.configs(connection_parameters).create()

Streamlit in Snowflake

Streamlit in Snowflake lets you build, deploy and share Streamlit apps on Snowflake without moving data out. Access is controlled with RBAC (role-based access control, where privileges go to roles and roles go to users), and apps run on a warehouse runtime or a container runtime (Snowflake documentation, October 2026). Teams use it for internal dashboards, data entry screens and small AI apps. It is a good portfolio project for a fresher who knows a little Python.

4. AI tools: Cortex, Snowflake CoWork and MCP

Short answer: Cortex is Snowflake's family of AI features, Snowflake CoWork (formerly Snowflake Intelligence) is the business-user chat product built on them, and the managed MCP server lets outside AI clients use Snowflake data safely.

Cortex is Snowflake's umbrella name for AI inside the platform. The main tools, with status as Snowflake states it (Snowflake documentation and release notes, October 2026):

MCP, the Model Context Protocol, is an open standard that lets AI assistants call outside tools in a consistent way. Snowflake's managed MCP server is the supported option. The older community repo under Snowflake-Labs is deprecated and should not be used for new work. Our dedicated Snowflake MCP server post covers setup, and our guide to Cortex AI functions has a hands-on lab.

Three cautions for October 2026:

5. Ingestion and ETL/ELT tools: getting data into Snowflake

Short answer: For files, use a stage with COPY INTO for batch loads and Snowpipe for continuous loads; for rows and events, use Snowpipe Streaming or the Kafka connector; for SaaS apps and databases, use Openflow or a third-party ELT tool such as Fivetran, Airbyte, Informatica or Matillion.

Native loading tools

Snowflake Openflow

Snowflake Openflow is Snowflake's own data integration service, built on Apache NiFi. It comes in two deployment options: a Snowflake deployment that runs on Snowpark Container Services (GA in AWS, Azure and Google Cloud commercial regions) and BYOC, where the data plane runs in your own cloud account (GA in AWS commercial regions). Its connectors include Google Drive, Box, SharePoint, change data capture (CDC) from MySQL, PostgreSQL, Oracle, SQL Server and MongoDB, Kafka, Kinesis, Salesforce, Google Ads, LinkedIn Ads and Shopify (Snowflake documentation, October 2026). Openflow is not enabled on trial accounts by default. Our Snowflake Openflow guide covers its architecture and setup.

Third-party ETL and ELT tools

Many teams use a separate ingestion tool, especially when they already have one. These tools are commonly used with Snowflake; we have not checked partner tiers or certifications, so confirm details on each vendor's own documentation.

Which Snowflake tool is recommended for batch data processing?

This is a common exam-style question. For batch loads of files, the native answer is COPY INTO from a stage, run on a virtual warehouse and usually scheduled with a task. Snowpipe is designed for continuous, file-by-file loading, and Snowpipe Streaming is for row-level real-time data. Use this simple rule:

6. Transformation tools: SQL, dbt and Snowflake's built-in options

Short answer: Most transformation on Snowflake is SQL, organised either with Snowflake objects such as streams, tasks, dynamic tables and stored procedures, or with dbt, which is the most requested companion tool in Indian Snowflake job descriptions.

Built-in transformation tools

dbt

dbt (data build tool) is an open-source tool that lets you write transformations as SELECT statements in version-controlled files, test them and build them in dependency order. It does the "T" in ELT. In the 14 Snowflake JDs we read on 25 October 2026, dbt was named in 9, and many titles now read "Snowflake + DBT". A Naukri search for "dbt snowflake" showed about 1,729 results on the same day. These are keyword-match counts, not a number of open jobs.

You can run dbt in three ways:

Is dbt required for Snowflake? No. Snowflake works fully without it. But the most detailed JD we read, from Capgemini, asked for modular dbt models, dbt tests, snapshots for SCD Type 2 (slowly changing dimensions that keep history) and two or more years of production dbt. If you are aiming at Snowflake data engineering roles, dbt is the tool to learn right after SQL.

7. Orchestration tools: tasks, Airflow and others

Short answer: Snowflake tasks schedule SQL and procedures inside Snowflake; Apache Airflow and similar tools orchestrate pipelines that span many systems.

Tasks and task graphs

A task runs a SQL statement or calls a stored procedure on a schedule, or when a stream has new data. Serverless tasks let Snowflake manage the compute; user-managed tasks name a warehouse. Task graphs chain tasks in series or in parallel. A new task starts suspended, so you must resume it (Snowflake documentation, October 2026). Here is the documented example, slightly expanded:

-- simplified example: a scheduled task
CREATE TASK SCHEDULED_T1
  WAREHOUSE = 'COMPUTE_WH'
  SCHEDULE = '60 MINUTES'
  AS SELECT 1;

ALTER TASK SCHEDULED_T1 RESUME;

Streams and tasks together give you incremental pipelines without any outside tool. Our streams and tasks lab builds one with MERGE.

Apache Airflow and other orchestrators

Apache Airflow is an open-source orchestrator that runs pipelines written in Python as DAGs (directed acyclic graphs of steps). It is commonly used with Snowflake through the Python connector. In our October 2026 JD sample, Airflow appeared in Capgemini and EPAM JDs and as "good to have" in a Cognizant one. Capgemini also listed Dagster and dbt Cloud as orchestration options.

8. BI and reporting tools

Short answer: Tableau and Power BI are the BI tools most often named alongside Snowflake in the JDs we read; Streamlit in Snowflake and Snowflake CoWork cover apps and natural-language questions inside Snowflake itself.

BI tools connect through the ODBC or JDBC drivers, send SQL to a Snowflake warehouse and show the results as dashboards. In our JD sample, Deloitte named Tableau and Power BI, and Caterpillar listed Power BI as preferred. Both are commonly used with Snowflake; check each vendor's own documentation for connector details.

Three things matter when you connect any BI tool:

Inside Snowflake, Streamlit in Snowflake is a good choice for custom data apps, and Snowflake CoWork (formerly Snowflake Intelligence) lets business users ask questions in plain language. Our Snowflake Intelligence guide explains how it works.

9. DevOps and infrastructure-as-code tools

Short answer: Teams manage Snowflake objects as code with the Terraform provider, keep SQL and dbt code in Git, and deploy through CI/CD tools such as GitHub Actions or Azure DevOps, often calling Snowflake CLI.

10. Monitoring, cost and governance tools

Short answer: Snowflake's built-in cost and monitoring tools are resource monitors for warehouses, budgets for serverless and AI spend, ACCOUNT_USAGE views for history, and Horizon Catalog for governance.

Credits per hour double at each warehouse size, from 1 for X-Small to 512 for 6X-Large, and warehouses bill per second with a 60-second minimum each time they start (Snowflake documentation, October 2026). In Snowflake's pricing table effective 15 October 2026, on-demand credits in AWS US East cost $2, $3, $4 and $6 by edition (Standard, Enterprise, Business Critical, VPS), and storage costs $23 per TB per month. Prices vary by region and edition, so check your own contract.

Here is the documented resource monitor example:

-- simplified example: cap a warehouse at 1,000 credits
CREATE OR REPLACE RESOURCE MONITOR limit1
  WITH CREDIT_QUOTA = 1000
  TRIGGERS ON 100 PERCENT DO SUSPEND;

ALTER WAREHOUSE wh1 SET RESOURCE_MONITOR = limit1;

Common ACCOUNT_USAGE views include QUERY_HISTORY, WAREHOUSE_METERING_HISTORY, LOGIN_HISTORY and STORAGE_USAGE, all in the SNOWFLAKE database. For hands-on tuning, read our query optimization guide. Third-party observability, data quality and cost tools also exist; evaluate them only after you have used the native tools, because they read the same metadata.

11. Which Snowflake tools Indian job descriptions ask for

Short answer: In the 14 Snowflake JDs we read on 25 October 2026, SQL appeared in every one, dbt and Python in 9 each, and Git or CI/CD in 5; specific ETL and BI products appeared far less often.

We read 14 Snowflake job descriptions from Naukri and LinkedIn on 25 October 2026, from companies such as Infosys, TCS, HCLTech, Capgemini, EPAM, Deloitte, DTCC and Caterpillar (listings seen on 25 Oct 2026). Here is how often each tool was named:

Two more points from the same sample. Most JDs asked for AWS or Azure alongside Snowflake. And none accepted 0 years of experience; the lowest ask was 2 to 5 years. For context, a Naukri search for "snowflake" returned about 25,910 results across India on 25 October 2026, including about 4,969 in Hyderabad and 8,463 in Bengaluru. A search for "snowpark" returned only 175, because Snowpark usually appears as a skill, not a title. All of these are keyword-match counts, not open jobs.

The lesson: tool lists in JDs are long, but the core is stable. SQL, Snowflake features, dbt, Python and one cloud get you shortlisted. A specific ETL or BI product is usually a bonus you can learn on the job.

12. Snowflake tools by role

Short answer: Developers live in SQL, Snowsight and dbt; data engineers add ingestion, orchestration and Python; admins focus on roles, warehouses, cost and automation; analysts focus on SQL and BI.

For full role guides, read our posts on the Snowflake developer and Snowflake data engineer roles.

13. How to choose and learn Snowflake tools

Short answer: Learn native tools before partner tools, pick one tool per layer, and let your background decide where you start.

Use this checklist when you pick tools for a project or for your own learning:

Where to start, by background:

Common mistakes to avoid:

14. How Brolly Academy helps

Short answer: Brolly Academy's 45-day Snowflake training in Hyderabad teaches the core native tools hands-on and introduces several companion tools from its tools list.

The course page lists these tools: Snowflake and the Snowflake web UI, SnowSQL CLI, Snowpipe, SQL, the Python connector, dbt, Power BI, Tableau, DBeaver, SQL Workbench, AWS S3, Azure Blob Storage and Google Cloud Storage. Its 15 modules cover architecture, data loading with internal and external stages and COPY INTO, Snowpipe with PUT and GET, Partner Connect (Python and BI connectors), JSON and semi-structured data, scheduling with tasks and streams, secure data sharing, cloning, Time Travel and Fail-safe, caching and performance, and advanced topics such as resource monitoring, data masking and Python integration through the Snowflake Python connector. Note that dbt appears in the tools list, not as a separate syllabus module.

Frequently asked questions