Rustem Galiullin
Research Engineer at Analog AI
Abu Dhabi Emirate, United Arab Emirates
Summary
👤
Senior🎓
Top SchoolRustem Galiullin is a research engineer with 9 years’ experience building and deploying deep learning and ML systems across computer vision, NLP, and remote sensing, currently engineering AI solutions at Analog AI in Abu Dhabi. He has driven production-grade inference platforms and automated training pipelines—cutting inference throughput by 22% at Bayanat—and implemented serverless Triton deployments for streaming CV at G42. An active open-source contributor, Rustem added CLIP zero-shot prediction, embeddings hooks, and unit tests to the popular FiftyOne project to improve dataset refinement and model workflows. Trained originally in accounting, he combines analytical rigor with hands-on ML research, often bridging prototype research and reliable production deployment.
9 years of coding experience
7 years of employment as a software developer
Bachelor’s Degree, Accounting and Business/Management, excellent, Bachelor’s Degree, Accounting and Business/Management, excellent at Kazan State UniversityMaster's degree, Accounting and ERP, 4, Master's degree, Accounting and ERP, 4 at University of Arkansas
English, Russian, Tatar, Chinese
Github Skills (11)
computer-vision10
pytorch10
machine-learning10
deep-learning10
clip10
python10
data-science9
developer-tools8
image-classification7
active-learning7
unstructured-data6
Programming languages (10)
Github contributions (5)
voxel51/fiftyone
Apr 2022 - Aug 2022
Refine high-quality datasets and visual AI models
Role in this project:
ML Engineer
ML EngineerContributions:13 reviews, 7 commits, 12 PRs in 4 months
Contributions summary:Rustem contributed to the integration of the CLIP model for zero-shot predictions, implementing functionalities to make predictions using default values. They updated documentation and examples for zero-shot usage, including custom class labels and text prompts. The user also added an embeddings hook for the CLIP model and wrote unit tests for the model zoo. They refactored the CLIP utilities into a subpackage.
pytorchpythonvisiondata-sciencedeep-learning
Rusteam/rusteam.github.io
May 2020 - May 2023
Blog at https://rusteam.github.io
Contributions:14 PRs, 20 pushes, 2 branches in 3 years
rustjekyllruby
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.