Andrei Tatarinov
Developer Technology Engineer at NVIDIA
California, United States
Summary
🤩
Rockstar🎓
Top SchoolAndrei Tatarinov is a Developer Technology Engineer with eight years of experience specializing in high-performance graph analytics and GPU-accelerated algorithms. Based in California, he combines deep C++ template metaprogramming expertise with CUDA-backed parallel implementations, contributing to prominent open-source RAPIDS projects like cuGraph, RAFT and cuML. His work has modernized spectral clustering and graph primitives—integrating RAPIDS Memory Manager hooks and replacing legacy NVGRAPH paths—so production ML pipelines gain both performance and cleaner dependencies. Trained in mathematics and computer graphics at Lomonosov MSU, he brings a rigorous, research-informed approach to engineering that balances low-level optimization with practical library integration.
8 years of coding experience
Lomonosov Moscow State University
Stackoverflow
1reputation
58reached
1answer
1question
Github Skills (24)
algorithm10
algorithms10
graph-algorithms10
c-language10
graph10
machine-learning10
machine-learning-algorithms10
data-structure10
rapids10
gpu10
cuda10
data-structures10
raft10
cprogramming-language10
cuml10
Programming languages (4)
Github contributions (5)
rapidsai/cugraph
Feb 2019 - Mar 2022
cuGraph - RAPIDS Graph Analytics Library
Role in this project:
Backend Developer
Backend DeveloperContributions:333 reviews, 162 commits, 32 PRs in 3 years 2 months
Contributions summary:Andrei focused on implementing and integrating RMM (RAPIDS Memory Manager) hooks into the cuGraph library. They connected RMM hooks and made changes to the `rmm_utils.h` header file to enable RMM features. Additionally, the user worked on integrating weighted Jaccard bindings and created related test files, indicating contributions to the graph analytics functionalities, specifically around the weighted Jaccard algorithm.
cudagraph-analysisanalyticsgraph-analyticsgraphml
rapidsai/raft
May 2020 - Aug 2020
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:106 commits, 5 PRs, 91 comments in 2 months
Contributions summary:Andrei's commits primarily focus on the implementation of spectral graph partitioning algorithms within the context of the RAFT library. The code changes reveal the creation of a spectral partition header file and the addition of core functionality including the creation of kmeans and the main functionality. These changes indicate a focus on building fundamental algorithms for machine learning tasks, including those related to the repository's topics of nearest neighbors and vector search. The commits involve CUDA accelerated implementations for high-performance computing.
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.