Reza Safi
Senior Software Engineer at Roblox
Palo Alto, California, United States
Summary
👤
Senior🎓
Top SchoolReza Safi is a Senior Software Engineer in Palo Alto with nine years of experience building large-scale, distributed systems using Java, C++, C, and SQL, blending research rigor from a PhD with production engineering. He contributed to core Apache Spark backend improvements at Cloudera—fixing commit protocols, adding memory metrics, and reducing library exposure—which highlights his impact on a high-profile open-source analytics engine. At Roblox he works on the data platform, applying expertise in multi-threaded and event-based systems to production data pipelines. His academic work includes novel static analysis techniques for performance bugs and event races, showing a rare combination of program analysis research and hands-on backend development. Colleagues can count on him for durable system design, thoughtful diagnostics, and pragmatic fixes that scale.
8 years of coding experience
10 years of employment as a software developer
Doctor of Philosophy (Ph.D.), Computer Science, Doctor of Philosophy (Ph.D.), Computer Science at University of Southern CaliforniaBachelor of Science (BSc), Computer Engineering, Bachelor of Science (BSc), Computer Engineering at Iran University of Science and TechnologyMaster of Science (M.Sc.), Information Technology, Master of Science (M.Sc.), Information Technology at Sharif University of Technology
Github Skills (10)
javas10
big-data10
spark10
java10
scala10
hadoop9
kafka9
sql8
data-pipeline8
data-pipelines8
Programming languages (4)
Github contributions (5)
apache/spark
Dec 2017 - Dec 2017
Apache Spark - A unified analytics engine for large-scale data processing
Role in this project:
Back-end Developer
Back-end DeveloperContributions:1 review, 1 commit, 15 PRs in 1 day
Contributions summary:Reza primarily focused on enhancing the Apache Spark backend, addressing issues related to Guava library exposure and improving the RDD commit protocol. They introduced new methods in `SessionCatalog` to remove exposures to the Guava library and ensured consistent JobIDs across executors and the driver during the Hadoop commit process. Furthermore, they contributed to giving users options for specifying a minimum message count per partition per batch and addressed memory-related issues by adding memory metrics. These modifications involved code changes in various core Spark components, including data storage, execution, and streaming modules.
analyticspythondata-processingsqlapache
rezasafi/spark
May 2017 - Oct 2024
Mirror of Apache Spark
Contributions:2 PRs, 105 pushes, 13 branches in 7 years 6 months
spark-mlapachebig-datasparkscala
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.