Parquet files are binary — you can’t just open them in a text editor. PondPilot uses DuckDB WASM to let you browse, inspect, and query Parquet files directly in your browser.
Instant Schema Inspection
Drop a Parquet file into PondPilot and immediately see:
- Column names and data types
- Row count
- Parquet metadata (row groups, compression)
SELECT * FROM parquet_metadata('dataset.parquet');
No pyarrow, no parquet-tools, no command line. Just drag and drop.
Query with Predicate Pushdown
DuckDB reads Parquet files intelligently. When you filter or select specific columns, DuckDB skips irrelevant row groups and columns:
SELECT user_id, event_type, timestamp
FROM events.parquet
WHERE event_type = 'purchase' AND timestamp > '2024-01-01';
This query only reads the columns and row groups it needs — fast even on large files.
Browse Data Visually
PondPilot renders query results in a sortable grid. Quickly scan through Parquet data, spot patterns, and drill down with more specific queries.
Common Parquet Workflows
Data lake exploration: Got a Parquet export from your data lake? Explore it locally before writing pipeline code.
Validating exports: Check that a Parquet file has the expected schema, row count, and data distribution.
Format conversion: Query Parquet data and export results as CSV for colleagues who need a spreadsheet.
Comparing versions: Open two Parquet files and use EXCEPT or JOIN to find differences.
No Installation, No Uploads
PondPilot runs DuckDB WASM in your browser. Your Parquet files stay on your machine. Open app.pondpilot.io and start viewing Parquet files immediately.
Open Source
PondPilot is free and open source. No limits on file size (beyond your browser’s memory), no feature gates, no accounts.
Open PondPilot and drop your first Parquet file.