Julien Le Dem
Berkeley, California, United States
Summary
🤩
RockstarJulien Le Dem is a Principal Engineer based in Berkeley with 14 years of experience building foundational data infrastructure. He co-created Apache Parquet, is an Apache Arrow PMC member and leads the OpenLineage project, shaping columnar formats, in-memory representations and lineage standards used across the data ecosystem. At companies from Twitter to Datadog and Astronomer he combines hands-on systems work—file formats, IPC, dictionary encoding and metadata—with architecture and team leadership. His open-source contributions span low-level format design (Parquet, Arrow) to operational tooling and testing improvements (Drill, Marquez, OpenLineage), and include pragmatic engineering such as ensuring Java/C++ interoperability and adding Parquet metadata/versioning. That blend of protocol-level design and production-focused tooling makes him a rare engineer who moves projects from spec to widely adopted implementations.
15 years of coding experience
Github Skills (73)
testing10
metadata10
file-format10
scala210
apache-arrow10
openapi-310
javas10
encoding10
parquet10
openapi10
schema10
client-library10
javascript10
aggregator10
encodings10
Programming languages (12)
Github contributions (5)
apache/parquet-format
Jan 2013 - Jan 2020
Apache Parquet Format
Role in this project:
Back-end Developer
Back-end DeveloperContributions:15 reviews, 85 commits, 7 PRs in 7 years 1 month
Contributions summary:Julien primarily contributed to the Apache Parquet format definition files. Their work involved modifying the Thrift IDL files, adding new data types, and refactoring the dictionary encoding. They improved the metadata handling, added statistics, and implemented a new data page type. Additionally, the user added a utility class to hide Thrift and updated the changelog, demonstrating a strong understanding of the Parquet file format's internal structure and its evolution.
avroparquetapachebig-dataapache-parquet
apache/parquet-java
Jan 2015 - Aug 2024
Apache Parquet Java
Role in this project:Back-end Developer
Back-end DeveloperContributions:5 reviews, 11 PRs, 3 pushes in 9 years 9 months
Contributions summary:Julien's commits primarily focus on refactoring code to use indices instead of field names, streamlining the process of reading and writing data within the Parquet format. They further enhanced the data storage by integrating support for decimal values with the aid of the existing Binary object, while also integrating new values to the page level encoding. The user also addressed bugs associated with handling empty fields.
avroparquetapachebig-dataapache-parquet
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.