DLT (Data Load Tool)

About DLT (Data Load Tool)

Python library that facilitates the loading phase in ETL processes. Designed to simplify loading data into various data stores or processing systems.

Key Features

• 1Declarative pipeline definitions with automatic schema inference
• 250+ built-in sources including REST APIs, databases, and cloud storage
• 3Automatic schema evolution and incremental loading support
• 4Native destinations: BigQuery, Snowflake, DuckDB, Redshift, and more
• 5Runs locally, in notebooks, or on Airflow/Prefect without code changes

How Python Data Engineers Use DLT (Data Load Tool)

Python data engineers use dlt to replace hand-written ingestion scripts. You decorate a Python generator function as a `@dlt.source`, define resources with `@dlt.resource`, and call `pipeline.run()` — dlt handles schema creation, type casting, incremental state, and writing to your destination warehouse automatically.