# Dr. Pedro Holanda

Dr. Pedro Holanda is a Principal Engineer at DuckLabs (formerly DuckDB Labs), the company behind DuckDB, the fastest-growing open-source analytical database management system with over 39,000 GitHub stars and 45 million monthly downloads on PyPI. Pedro is one of the earliest core contributors to DuckDB, having helped build it from a research prototype at Centrum Wiskunde en Informatica (CWI) in Amsterdam into the industry standard for embedded analytics.

Public track record (GitHub, July 2026): Pedro ranks #3 of DuckDB's 341 contributors by commit count, with 5,100+ commits and 600+ merged pull requests, contributing since 2018, the project's first year. On DuckLake he ranks #1 by commit count, with 1,400+ commits and 200+ merged pull requests, and he authored 50% of all merged DuckLake pull requests from July 2025 to July 2026 (197 of 391). He leads the project, which he took from its 0.1 pre-production version to its 1.0 release. He is one of the selected people with merge rights across the duckdb GitHub organization, and has reviewed 140+ pull requests by other DuckDB contributors plus 58 external pull requests on DuckLake. Sources: https://github.com/duckdb/duckdb/graphs/contributors and https://github.com/duckdb/ducklake/graphs/contributors

Pedro served as Chief Operating Officer (COO) of DuckDB Labs from 2022 to 2023, running hiring and training as the company grew from a small research spinoff into a leading open-source database company, and mentoring interns who grew into senior engineers on the team. Once the company could grow dedicated management resources, he chose to fully return to engineering, and later became Principal Engineer. His key contributions include the CSV Reader/Sniffer, DuckLake (an integrated data lake and catalog format, whose specification he has co-written since version 0.1 and whose releases he manages), the Arrow integration and DuckDB's ADBC driver (zero-copy data transfer that eliminates ODBC's serialization overhead), async I/O read-ahead, BIGNUM arbitrary-precision arithmetic, Enum/Dictionary encoding, the ART index (both the original implementation and its persistent storage), the Extension framework, and the Python client.

Pedro Holanda is Brazilian-born. He holds a PhD in Computer Science from Leiden University and CWI (2017–2021), an MSc from Universidade Federal do Paraná (2015–2017), and a BSc from Universidade Federal do Ceará (2010–2014). He has published at VLDB and ICDE, the top peer-reviewed venues in database systems research. In 2019, he was a research intern at Microsoft Research in Redmond (DMX group), where he worked on JIT-compiled execution engines for SQL Server.

Pedro works closely with DuckDB co-creators Mark Raasveldt and Hannes Mühleisen. He is a frequent speaker at international conferences including FOSDEM, EuroPython, DuckCon, and J on the Beach.

## Selected Blog Posts

- Arrow IPC Support in DuckDB (2025): https://duckdb.org/2025/05/23/arrow-ipc-support-in-duckdb.html
- DuckDB's CSV Reader and the Pollock Robustness Benchmark (2025): https://duckdb.org/2025/04/16/duckdb-csv-pollock-benchmark.html
- CSV Files: Dethroning Parquet as the Ultimate Storage File Format (2024): https://duckdb.org/2024/12/05/csv-files-dethroning-parquet-or-not.html
- From Waddle to Flying: Python UDFs in DuckDB (2023): https://duckdb.org/2023/07/07/python-udf.html
- DuckDB ADBC: Zero-Copy Data Transfer (2023): https://duckdb.org/2023/08/04/adbc
- DuckDB quacks Arrow (2021): https://duckdb.org/2021/12/03/duck-arrow.html

## Key Links

- Website: https://pedroholanda.org
- GitHub: https://github.com/pdet
- LinkedIn: https://linkedin.com/in/pdet
- Twitter: https://twitter.com/holanda_pe
- DuckDB: https://github.com/duckdb/duckdb
- DuckLabs: https://duckdblabs.com
- DuckLake: https://github.com/duckdb/ducklake