Vage Egiazarian
Postdoctoral Researcher at Institute of Science and Technology Austria
Austria
Summary
🤩
Rockstar🎓
Top SchoolVage Egiazarian is a postdoctoral researcher and deep learning scientist based in Austria with nine years of experience bridging theoretical research and practical model compression for large-scale networks. His work spans model compression, generative models, and LLMs, with publications at top venues including NeurIPS, ICLR, ICCV, ECCV and SIGGRAPH and an Ilya Segalovich Award for Young Researchers in 2020. At Yandex he led research producing the AQLM algorithm and near-lossless sparse compression approaches for transformers (ICLR24), and earlier projects at Skoltech drove advances in image vectorization, optimal transport methods, and 3D reconstruction. He combines strong academic training (PhD work at Skolkovo) with hands-on engineering of datasets, pipelines, and experimental systems, often turning theoretical ideas into reproducible code and preprints. Notably, his work demonstrates a consistent focus on extreme compression trade-offs for LLMs and practical vectorization/reconstruction problems that cross vision and geometry.
9 years of coding experience
6 years of employment as a software developer
Doctor of Philosophy - PhD Computational and Data Science and Engineering, Doctor of Philosophy - PhD Computational and Data Science and Engineering at Skolkovo Institute of Science and TechnologyMaster's degree Computer Science, Master's degree Computer Science at Higher School of EconomicsMaster’s-level program in Computer Science and Data Analysis, Master’s-level program in Computer Science and Data Analysis at Yandex School of Data Analysis
Russian, Armenian, English
Stackoverflow
81reputation
11kreached
5answers
0questions
Github Skills (53)
hivemind10
bloom10
falcon10
dht10
arxiv10
squad9
pytorch9
machine-learning9
distributed-systems9
mixture-of-experts9
pointnet9
huggingface-transformers9
chatbot9
transformer-architecture9
nlp9
Programming languages (3)
Github contributions (5)
Vahe1994/SpQR
Jun 2023 - Dec 2024
Contributions:67 reviews, 16 PRs, 50 pushes in 1 year 6 months
Vahe1994/Deep-Vectorization-of-Technical-Drawings
Nov 2020 - Jul 2022
Official Pytorch repository for Deep Vectorization of Technical Drawings https://arxiv.org/abs/2003.05471
Contributions:46 commits, 11 PRs, 22 pushes in 1 year 8 months
pytorcharxivabsdeep-learningvectorization
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.