Divye Gala
Senior Software Engineer at NVIDIA
New York, New York, United States
Summary
🤩
Rockstar🎓
Top SchoolDivye Gala is a Senior Software Engineer with a decade of experience building high-performance, GPU-accelerated systems for large-scale machine learning and graph analytics. Based in New York and contributing to NVIDIA RAPIDS, he has shipped core improvements to cuDF, cuML, and RAFT—optimizing GPU DataFrame operations, accelerating K-Means and HDBSCAN, and implementing a fast Boruvka MST that yielded up to 48x speedups on RMAT graphs. He blends low-level CUDA and C++ engineering with practical ML expertise, including sparse-matrix pairwise metrics using shared-memory hash tables and multi-GPU dataset generators for on-GPU testing. An active open-source maintainer, Divye focuses on correctness and performance of foundational analytics primitives that enable real-time inferencing and vector search at scale.
10 years of coding experience
1 year of employment as a software developer
Master of Science - MS, CSE, Master of Science - MS, CSE at Georgia Institute of TechnologyBachelor of Engineering, Computer Engineering, Bachelor of Engineering, Computer Engineering at Bhartiya Vidya Bhavans Sardar Patel Institute of Technology Munshi Nagar Andheri Mumbai
Stackoverflow
36reputation
1kreached
1answer
1question
Github Skills (30)
algorithm10
algorithms10
c-language10
k-means-clustering10
dataframes10
gpu-programming10
machine-learning10
cudf10
building-blocks10
data-structure10
dataframe10
gpu10
cuda10
data-structures10
cprogramming-language10
Programming languages (9)
Github contributions (5)
rapidsai/cuml
Jun 2019 - Jan 2023
cuML - RAPIDS Machine Learning Library
Role in this project:
ML Engineer
ML EngineerContributions:197 reviews, 526 commits, 130 PRs in 3 years 7 months
Contributions summary:Divye's commits primarily focus on the development and improvement of the K-Means clustering algorithm within the cuML library. The user modifies and extends the API and implementations of core functions, enhancing the functionality and performance of the algorithm. Their work demonstrates proficiency in data science and machine learning, particularly in the context of GPU-accelerated algorithms.
cudacumlnvidiadata-sciencegpu
rapidsai/raft
Jul 2020 - Nov 2022
RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing high performance applications.
Role in this project:Back-end Developer
Back-end DeveloperContributions:483 reviews, 60 commits, 104 PRs in 2 years 4 months
Contributions summary:Divye primarily focused on modifying and refactoring the CUDA-accelerated algorithms and primitives within the repository. Their work included removing and re-adding include files, and making changes to handle and comms, indicating work in low-level system/library code. The user was also adding new features, adding get alloc to the handle, and adding tests, which showcases a development of the code base. The changes also included fixing typos and debugging code, further contributing to the stability and maintenance of the codebase.
sciencedata-sciencemachine-learninggraph-data-sciencefundamental
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.