What are the main use cases for DuckDB?

DuckDB's combination of speed, simplicity, and direct file-querying capability suits a range of scenarios where a full server-based data warehouse would be excessive overhead but real analytical SQL power is still needed.

1. Exploratory data analysis - quickly querying local CSV/Parquet files or a Pandas DataFrame without setting up infrastructure.
2. Embedded analytics - powering analytical features directly inside an application, without a separate database service to operate.
3. ETL/ELT and data pipeline steps - transforming data efficiently as a lightweight processing step, often faster than the equivalent Pandas-only or Spark-only approach for moderate data sizes.
4. Querying data lakes directly - running SQL against Parquet/Iceberg/Delta Lake/DuckLake files in object storage without loading them into a separate warehouse first.
5. In-browser analytics - via DuckDB-Wasm, for client-side, no-backend-required data applications.
6. Teaching and prototyping - a fast, dependency-free way to explore SQL and analytical concepts without database administration overhead.

The unifying thread is scenarios where DuckDB's zero-infrastructure, single-process model is a genuine advantage rather than a limitation, cases where standing up and operating a traditional client-server data warehouse would be disproportionate to the actual data volume or query complexity involved.