Alexander Kozlov
Principal Engineer at Cerebras Systems
Dubai, Dubai, United Arab Emirates
Summary
🤩
Rockstar🎓
Top SchoolAlexander Kozlov is a Principal Engineer based in Dubai with over eight years of experience specializing in deep learning research, low-precision architectures, and machine learning systems engineering. He has driven ML and computer vision work across Intel and Cerebras, with deep expertise in model compression, quantization, and production-ready inference optimizations. Alexander contributes to the high-profile OpenVINO open-source toolkit, moving post-training optimization tools into OSS, improving LLM compression docs, and building accuracy tests for large models—demonstrating a rare blend of research rigor and engineering pragmatism. He holds a PhD in physics and mathematics and a top-ranked master’s in applied mathematics, reflecting strong theoretical foundations that he applies to practical system and performance problems. Known for bridging low-level system programming and higher-level ML toolchains, he brings both hands-on implementation skills (C/Linux frameworks) and ML framework fluency (Caffe, TensorFlow, PyTorch) to scale AI inference.
8 years of coding experience
16 years of employment as a software developer
Doctor of Philosophy - PhD, physics and mathematics, Doctor of Philosophy - PhD, physics and mathematics at Нижегородский Государственный Университет им. Н.И. Лобачевского (ННГУ)
Русский, Английский
Github Skills (9)
mlops10
deep-learning10
ai10
openvino10
model-optimization10
python9
generative-ai9
computer-vision8
pytorch8
Programming languages (3)
Github contributions (5)
openvinotoolkit/openvino
Oct 2021 - Jan 2023
OpenVINO™ is an open-source toolkit for optimizing and deploying AI inference
Role in this project:
ML Engineer
ML EngineerContributions:481 reviews, 11 commits, 180 PRs in 1 year 3 months
Contributions summary:Alexander's contributions primarily involve the development and optimization of machine-learning related tools within the OpenVINO toolkit. This includes moving post-training optimization tools to open-source, fixing dependencies, and implementing documentation for NNCF PTQ and weight compression. The user has also added a Gen AI landing page and updated LLM compression-related information, demonstrating knowledge in the domain. Furthermore, the user's work encompassed creating and updating accuracy tests for LLMs, showcasing a focus on model evaluation and performance.
inference-enginepytorchmodel-optimizerdeep-learninggpu
AlexKoff88/nncf_pytorch
Sep 2020 - Feb 2024
Neural Network Compression Framework for PyTorch*
Contributions:53 pushes, 14 branches in 3 years 5 months
pytorchneural-network-compressiondeep-learningcompressionneural-network
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.