Arrow and Feather files are everywhere in the modern data stack — fast, columnar, language-neutral. They’re also annoying to inspect without booting up Python or R. PondPilot opens .arrow and .feather files directly in your browser.
Open Any Arrow or Feather File
- Open app.pondpilot.io
- Drop in your
.arrowor.featherfile - Instantly see the schema and a data preview
- Query with SQL
The file is read in place via DuckDB WebAssembly. No conversion, no import step, no upload.
Zero-Copy, Fast
Arrow’s whole point is zero-copy columnar data. DuckDB speaks Arrow natively, which means reads are efficient — the engine and the file format share the same memory layout. Large Arrow files open quickly because DuckDB only materializes the columns your query touches.
Full SQL on Top
Most Arrow viewers show you a table and stop there. PondPilot gives you DuckDB’s full analytical SQL:
SELECT
date_trunc('day', event_ts) AS day,
event_type,
COUNT(*) AS n
FROM 'events.arrow'
GROUP BY 1, 2
ORDER BY 1 DESC, n DESC;
Aggregations, joins across files, window functions — all the things a plain viewer can’t do.
Feather v1 and v2
Feather started as a simple on-disk Arrow format and evolved into Feather v2 (essentially Arrow IPC). PondPilot handles both. If the file opens in PyArrow or R’s arrow package, it opens here.
Schema Inspection
Arrow files carry full type information — nested structs, lists, dictionaries, extension types. The schema panel shows all of it, which is invaluable when you receive a file and need to understand its shape before writing queries.
Export to Other Formats
View Arrow, export Parquet, CSV, JSON, or Excel. Handy for sharing a slice with someone who doesn’t live in the Arrow ecosystem.
Privacy
Arrow files are often intermediate artifacts in data pipelines — they can contain anything. PondPilot processes everything in-browser. No server sees your file.
Get Started
Visit app.pondpilot.io and open your Arrow or Feather file.