Hugo L.
Sample My laptop is now my warehouse
I analyse tens of millions of rows of public transport data that used to crash pandas. The file querying lessons were exactly what I needed.
DuckDB for analysts: fast analytics on your own laptop
Query CSV, Parquet and JSON files with SQL, explore millions of rows in seconds and share results from Python, without a server or a warehouse
Created by Priya Raghavan
- Updated September 2026
- English
- Intermediate
What you will learn
- Query CSV, Parquet and JSON files with SQL without loading them first
- Profile a new dataset and find quality problems in minutes
- Use DuckDB SQL extensions to write shorter, clearer queries
- Read remote Parquet files from object storage over HTTPS or S3
- Combine DuckDB with pandas or Polars in Python workflows
- Export clean results to Parquet, CSV or Excel for others
Course content
Getting productive with DuckDB
Cleaning and shaping data
- Video: Fixing types, dates and text 10 min
- Video: Handling duplicates and bad rows 9 min
- Article: Joining files from different sources 11 min
- Video: Nested data: structs, lists and maps 10 min
Analysis patterns
- Video: Time series with date functions and windows 11 min
- Article: Approximate and statistical functions 9 min
- Video: Top N, percentiles and outliers 10 min
- Video: Geospatial basics with the spatial extension 10 min
Remote data and extensions
- Article: Reading files over HTTPS and from S3 11 min
- Video: Attaching other databases 9 min
- Video: Writing results for others 9 min
Performance and limits
- Article: How DuckDB uses memory and disk 9 min
- Video: Reading query plans with EXPLAIN ANALYZE 10 min
- Video: When to move beyond a laptop 8 min
DuckDB from Python
- Article: The Python API and relations 10 min
- Video: Working alongside pandas and Polars 11 min
- Video: Reproducible analysis projects 9 min
- Article: Sharing results: notebooks, apps and the DuckDB UI 9 min
- Quiz: Final check 6 min
Requirements
- Basic SQL: SELECT, WHERE, GROUP BY, joins
- Some Python is helpful for the later sections
Description
DuckDB is an analytical database that runs inside the process that uses it: in your terminal, in a Python script or in a notebook. There is no server to install and no account to create, and it can scan millions of rows from Parquet or CSV files in seconds on an ordinary laptop. For analysts this changes what is practical to do locally.
What you will learn
- Querying files directly: CSV, Parquet, JSON, and folders of them
- DuckDB's friendly SQL additions such as GROUP BY ALL, EXCLUDE, SUMMARIZE and PIVOT
- Profiling unfamiliar data quickly and spotting problems early
- Joining local files with remote data in object storage
- Using DuckDB from Python alongside pandas and Polars
- Packaging an analysis so a colleague can rerun it exactly
Who it is for
You should already know basic SQL. The course is aimed at analysts who hit the limits of spreadsheets or pandas memory, and at engineers who want a fast local tool for exploring data before it reaches a warehouse.
We use DuckDB 1.x through its command line client and Python package. Datasets are public: city bike trips, taxi rides and open government data, ranging from a few thousand rows to tens of millions.
Who this course is for
- Analysts whose spreadsheets or pandas scripts have become slow
- Data engineers who want a fast local tool for exploring files
- Researchers working with large public datasets
Instructor
Priya Raghavan
Streaming and data platform engineer. Kafka, Python pipelines and DuckDB explained plainly
- 4.4 instructor rating
- 21 reviews
- 7,170 students
- 3 courses
- Sample
I work on data platforms where events arrive all day and someone always wants yesterday's numbers fixed. Most of my time goes into Kafka topics, Python ingestion services and the small analytical tools that make debugging them faster.
I started teaching because new data engineers kept asking me the same questions: why is this consumer lagging, why did this file load twice, why is this query slow. My courses answer those questions with small, complete examples rather than slides.
Outside work I maintain a few open source notebooks for teaching SQL over Parquet files, and I like explaining file formats more than I probably should.
Student reviews
Selin A.
Sample SUMMARIZE routine is brilliant
The twenty minute exploration routine is now my first step with every new dataset.
Eva R.
Sample Python integration is very useful
Querying pandas data frames by name with SQL replaced a lot of awkward merge code in my notebooks.
Kofi M.
Sample Great value for the price
Clear, short lessons. The spatial extension lesson was a nice surprise. I would like more on performance tuning.
Daniel K.
Sample Honest about limits
I appreciated the note that the friendly SQL shortcuts are DuckDB specific and where they differ in other warehouses.
Mira S.
Sample Good, wanted more on remote data
Solid course. The S3 and httpfs lesson was a bit brief for my use case, but the rest was very good.
Keep exploring
All Analytics and SQL courses
More in Analytics and SQL
$29.99
Enroll now