Arnab Phani
Postdoctoral Researcher at The Apache Software Foundation
Berlin, Germany
Summary
🤩
Rockstar🎓
Top SchoolArnab Phani is a research-focused software engineer and postdoctoral researcher in Berlin with 10 years of experience building scalable data and ML systems across academia and industry. He combines deep systems engineering from a long tenure at Teradata with academic rigor from a PhD track at TU Berlin and Graz, now advancing research and teaching as a postdoc. An active Apache SystemDS PMC member and contributor, he has implemented core backend features—such as matrix scaling, libsvm I/O support, performance tweaks to cbind, and lineage optimizations—showing strength in production-ready ML infrastructure. Comfortable moving between Java production code and research prototypes, he brings a pragmatic approach to optimizing end-to-end data science workflows. Colleagues value him for bridging open-source community work with reproducible research and measurable performance improvements.
10 years of coding experience
15 years of employment as a software developer
BITS Pilani, Birla Institute of Technology and ScienceBachelor of Technology (BTech) Information Technology, Bachelor of Technology (BTech) Information Technology at West Bengal University of Technology, KolkataDoctor of Philosophy - PhD, Doctor of Philosophy - PhD at Technische Universität Berlin
Github Skills (10)
javas10
mat10
matrix10
performance-optimization10
linear-algebra10
java10
spark9
data-structures8
scala8
data-structure8
Programming languages (3)
Github contributions (5)
apache/systemds
Apr 2019 - Jan 2023
An open source ML system for the end-to-end data science lifecycle
Role in this project:
Backend Developer
Backend DeveloperContributions:4 releases, 38 reviews, 182 commits in 3 years 9 months
Contributions summary:Arnab implemented a new `scale` built-in function, which scales and centers an input matrix, including the addition of test routines. This involved writing Java code within the `org.tugraz.sysds.test.functions.builtin` package. The user also added a new built-in function for time measurements. Additionally, the user added support for the libsvm input/output format and fixed a bug in the generation of lineage items for the `DataGenCPInstruction`. Furthermore, the user worked on enhancing the performance of cbind operations and implemented several partial rewrite and optimization steps for the lineage framework.
pythonend-to-enddmllifecycledata-science
phaniarnab/phaniarnab.github.io
Jan 2021 - Mar 2025
Github Pages template for academic personal websites, forked from mmistakes/minimal-mistakes
Contributions:141 pushes, 2 branches in 4 years 2 months
templategithub-pages-templatemmistakesmistakesjekyll
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.