Iceberg Browser Client — Query Iceberg Catalogs from Your Browser

What if you could browse and query your Iceberg tables from a browser tab? No Spark cluster, no Trino deployment, no JVM. Just open PondPilot, point it at your Iceberg REST Catalog, and start running SQL.

This is real — and it’s genuinely new.

How It Works

PondPilot runs DuckDB as WebAssembly in your browser. DuckDB’s Iceberg extension supports connecting to Iceberg REST Catalogs directly from WASM. That means your browser can read Iceberg table metadata, resolve snapshots, and query Parquet data files — all without a backend server.

Getting Started

  1. Open app.pondpilot.io
  2. Configure your Iceberg REST Catalog connection
  3. Browse namespaces and tables
  4. Run SQL queries against your Iceberg data
SELECT product_category, SUM(revenue) as total_revenue
FROM iceberg_catalog.sales.transactions
WHERE year = 2025
GROUP BY product_category

Why This Matters

Iceberg adoption is exploding, but querying Iceberg data still requires heavy infrastructure — Spark, Trino, or at minimum a running DuckDB process. PondPilot eliminates that for exploration and ad-hoc analysis. Open a browser tab and you’re querying production Iceberg tables.

Use Cases

  • Data exploration — browse what’s in your Iceberg catalog without spinning up compute
  • Quick validation — check that a pipeline wrote the right data
  • Demo and presentation — show Iceberg data live without provisioning infrastructure
  • Schema discovery — explore table schemas and partition layouts

No Infrastructure to Manage

There’s no server component. PondPilot connects directly from the browser to your Iceberg REST Catalog endpoint. You bring the catalog URL and credentials; PondPilot brings the SQL engine.

Get Started

Visit app.pondpilot.io and connect to your Iceberg catalog.

Back to home