How to Load CSV Data into DuckDB using dlt

📘 Introduction

CSV files are everywhere in data work. You might receive exports from business tools, download small datasets, or prepare local files before moving data into a warehouse.

In this tutorial, you will learn how to load CSV data into DuckDB using dlt. DuckDB gives us a fast local analytics database, and dlt helps us turn a simple file load into a repeatable data pipeline.

We will create a small CSV file, build a Python pipeline, load the file into DuckDB, and check the result with SQL.

💡 What are we implementing?

We will build a local pipeline with this flow:

CSV file -> dlt -> DuckDB table -> SQL check

The goal is not only to import a file once. The goal is to understand the basic pattern behind many data engineering workflows:

• define where the data comes from
• define where the data should go
• run the pipeline
• inspect the loaded table

For a beginner project, this is a great combination because everything runs locally. You do not need a cloud account, database server, or API key.

✅ Prerequisites

Before we start, you should have:

☑️ Python 3.9 or newer installed
☑️ Basic knowledge of running terminal commands
☑️ A text editor such as VS Code
☑️ No API key or cloud account

⚙️1️⃣ Create a project folder

First, create a new folder for the project:

mkdir dlt-csv-duckdb
cd dlt-csv-duckdb

Create and activate a virtual environment:

python -m venv .venv
source .venv/bin/activate

On Windows, activate it with:

.venv\Scripts\activate

📦2️⃣ Install package

Now install dlt with DuckDB support:

pip install "dlt[duckdb]"

This installs dlt together with the DuckDB Python package and the dependencies needed for the DuckDB destination. You do not need to install DuckDB separately for this tutorial.

📄3️⃣ Create a small CSV file

Create a folder named data:

mkdir data

Inside that folder, create a file named customers.csv:

customer_id,name,country,created_at
1,Ana Silva,Portugal,2026-01-10
2,John Miller,United States,2026-01-12
3,Mina Tanaka,Japan,2026-01-14
4,Sofia Garcia,Spain,2026-01-15

Your project should now look like this:

dlt-csv-duckdb/
├── .venv/
└── data/
    └── customers.csv

This public part already gives us a clean project setup and a realistic CSV file. In the Academy section, we continue by building the dlt pipeline, loading the file into DuckDB, checking the table with SQL, and fixing common beginner mistakes.