SQL Playground — Run SQL in Your Browser

Free, in-browser SQL engine powered by DuckDB-WASM. Write and execute SQL queries instantly on preloaded sample datasets — employees, orders, and web events. No server, no signup, no data leaves your device. Practice window functions, CTEs, QUALIFY, aggregations, joins, and more.

How It Works

This playground loads DuckDB-WASM, a full analytical SQL engine compiled to WebAssembly, directly in your browser. The engine is about 32 MB on the first visit, then cached by your browser. Three sample tables are preloaded with realistic data so you can start writing queries immediately.

Preloaded Sample Tables

• employees (15 rows) — id, name, department, salary, hire_date. Practice GROUP BY, HAVING, window functions, and top-N per group patterns.
• orders (15 rows) — order_id, customer_id, product, quantity, unit_price, order_date, region. Practice aggregations, running totals, and revenue analysis.
• web_events (15 rows) — event_id, user_id, event_type, page, ts, device. Practice funnel analysis, sessionization, and conditional aggregation.

What You Can Practice

Fundamentals

• SELECT, WHERE, GROUP BY, HAVING, ORDER BY
• JOINs (INNER, LEFT, RIGHT, FULL, CROSS)
• Subqueries and correlated subqueries
• UNION, INTERSECT, EXCEPT

Advanced SQL

• Window functions — ROW_NUMBER, RANK, DENSE_RANK, LAG, LEAD, SUM OVER, AVG OVER
• CTEs (WITH ... AS) and recursive CTEs
• QUALIFY clause (supported by DuckDB and Snowflake)
• CASE expressions, COALESCE, NULLIF

Data Engineering Patterns

• Running totals and moving averages with window frames
• Funnel analysis with conditional aggregation
• Sessionization using LAG and date arithmetic
• Deduplication with ROW_NUMBER

Interview Prep

• Top-N per group (a classic data engineer interview question)
• Year-over-year comparisons
• Percentile and median calculations
• Self-joins for sequential event analysis

DuckDB SQL Dialect

DuckDB supports a PostgreSQL-compatible dialect with modern extensions. Most Snowflake and BigQuery SQL patterns work with minor syntax adjustments. Supported features include: window functions, CTEs, recursive CTEs, QUALIFY, PIVOT/UNPIVOT, LATERAL joins, LIST/STRUCT/MAP types, regex functions, and lambda expressions.

Privacy and Security

DuckDB runs entirely in your browser via WebAssembly. Your SQL queries and any data you paste never leave your device — there is no server, no database connection, and no logging of queries. You can safely paste proprietary data for testing.

Frequently Asked Questions

Is this SQL playground free?

Yes, completely free. No signup, no limits, no tracking. DuckDB-WASM runs 100% in your browser.

What SQL dialect does DuckDB support?

DuckDB supports a PostgreSQL-compatible dialect with modern extensions: window functions, CTEs, QUALIFY, PIVOT/UNPIVOT, LATERAL joins, LIST/STRUCT/MAP types, regex, lambda functions, and more.

Can I load my own data?

The playground ships with three preloaded sample tables. You can also use DuckDB's read_csv() function to query CSV data directly.

How large a dataset can it handle?

DuckDB-WASM runs inside your browser's memory budget — typically 1-4 GB. For sample datasets (15 rows each), performance is instant. For analytical workloads up to ~100 MB, DuckDB-WASM performs well.

Does it support Snowflake-specific syntax?

DuckDB supports many Snowflake SQL patterns natively: QUALIFY, FLATTEN (as UNNEST), window functions, CTEs, MERGE, and most date/string functions. Snowflake-only functions (GET_DDL, SYSTEM$) are not available.

Related Tools

• SQL Formatter — clean up your SQL before running it here.
• Snowflake SQL Cheat Sheet — full Snowflake SQL reference with examples you can copy into the playground.
• Window Functions Reference — ROW_NUMBER, RANK, LAG, LEAD — all examples are runnable in this playground.
• Interview Prep Hub — complete 14-day study plan for data engineer interviews.

← Back to Home