Alexander Shirkov
Senior Software Engineer, Amazon Kuiper
Seattle, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolAlexander Shirkov is a senior software engineer with 11 years of experience building large-scale, production ML and distributed systems at Amazon and AWS, currently automating integrated vehicle testing for the Kuiper satellite program. He has deep expertise in ML tooling and deployment—contributing to high-profile open-source projects like AutoGluon, the SageMaker Python SDK, and AWS Deep Learning Containers to ensure compatibility, security, and GPU/CPU parity. His background spans backend architecture, data engineering, and DevOps, with earlier work at Deutsche Bank scaling event-processing systems and automating deployments across global regions. Alexander combines hands-on coding in Python, Java/Scala, and Spark with pragmatic system design to deliver reliable, observable services under tight deadlines. He cares about impactful projects that benefit many people, evidenced by shifts from financial systems to consumer-scale ML and satellite infrastructure. Collected contributions to AutoGluon and SageMaker show a knack for bridging research-grade models into robust, production-ready workflows.
11 years of coding experience
17 years of employment as a software developer
Master’s Degree Electro-Optical Equipment Design, Master’s Degree Electro-Optical Equipment Design at Belarusian State University of Informatics and Radioelectronics
Russian, English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (17)
pytorch10
docker10
pytest10
python10
data-science10
testing10
machine-learning10
datatable10
containerization10
dockers10
sagemaker10
data-preprocessing10
tensorflow10
aws10
tabular10
Programming languages (5)
Github contributions (5)
autogluon/autogluon
Sep 2020 - Jan 2023
Fast and Accurate ML in 3 Lines of Code
Role in this project:
ML Engineer & Data Scientist
ML Engineer & Data ScientistContributions:8 releases, 855 reviews, 135 commits in 2 years 5 months
Contributions summary:Alexander primarily contributed to the implementation of fast.ai tabular models within the AutoGluon framework, adding support for fast.ai's tabular neural network models. They expanded the data type selection to handle a wider range of data types, and addressed issues related to model training and inference, including handling single-row prediction failures. The user also made improvements to the data preprocessing steps and refactored code to handle `refit_full` functionality, indicating a focus on improving model performance and usability.
forecastingimage-textmlppythonmeta-learning
aws/deep-learning-containers
Aug 2021 - Jan 2023
AWS Deep Learning Containers are pre-built Docker images that make it easier to run popular deep learning frameworks and tools on AWS.
Role in this project:ML Engineer & DevOps Engineer
ML Engineer & DevOps EngineerContributions:107 reviews, 24 commits, 31 PRs in 1 year 5 months
Contributions summary:Alexander primarily contributed to the development and testing of training containers within the AWS Deep Learning Containers repository. Their work involved integrating Autogluon, adding training scripts, and implementing integration tests, including CPU vs. GPU testing. They also worked on container security fixes, ensuring dependencies were up-to-date, and addressed code review comments. Moreover, they were responsible for applying security recommendations to the containers.
pytorchsagemakercontainersmxnetserving
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.