Aleksandr Artemenkov
Verified Expert in Engineering
Machine Learning Engineer and Developer
Helsinki, Finland
Toptal member since December 20, 2021
Aleksandr is a machine learning engineer with a research background and extensive knowledge of essential DevOps tools. Specialized in end-to-end machine learning model development, he can conduct research, build scalable pipelines, and assist in their integration into the production environment. Aleksandr feels comfortable contributing to big companies and smaller projects within strong teams, always focusing on maintainability and quality results.
Portfolio
Experience
- Git - 5 years
- Data Science - 4 years
- PyTorch - 4 years
- Machine Learning - 4 years
- Python - 4 years
- C++ - 2 years
- Docker - 1 year
- PySpark - 1 year
Preferred Environment
Python, Git, Conda, Linux, PyTorch
The most amazing...
...package I've developed is a scalable visualization framework called NCVis that has more than 25,000 downloads on conda-forge, a GitHub organization.
Work Experience
Data Scientist
Ozon.ru
- Developed a streaming service for image feature extraction via a deep neural network.
- Reduced the pipeline memory consumption by more than 50% via implementing a custom PySpark daemon.
- Proposed a correction to the model validation process resulting in more than 10% error reduction in production.
- Implemented a ranking algorithm that increased the model's throughput by more than 20% while preserving the quality.
Research Intern
Skolkovo Institute of Science and Technology
- Published a paper at WWW ’20—a CORE rank A* conference—on noise contrastive dimensionality reduction.
- Developed a package for scalable visualization called NCVis that surpassed its competitors in terms of performance due to the OpenMP parallelization.
- Built a continuous development pipeline for the NCVis package using Azure Pipelines and conda-forge, resulting in more than 25,000 downloads to this date.
Experience
NCVis: Scalable Data Visualization
https://github.com/stat-ml/ncvisUncertainty Estimation Framework
https://github.com/alartum/sngp-pytorchIt is used for uncertainty estimation in deep neural networks and allows scalable inference. The implementation differs a bit from the original paper to achieve better performance in terms of speed.
Education
Master's Double Degree in Data Science and Applied Mathematics and Physics
Skolkovo Institute of Science and Technology | Moscow Institute of Physics and Technology - Moscow, Russia
Bachelor's Degree in Applied Mathematics and Physics
Moscow Institute of Physics and Technology - Moscow, Russia
Skills
Libraries/APIs
NumPy, PyTorch, Scikit-learn, OpenMP, PySpark, Spark ML, MPI, PyTorch Lightning, OpenCV
Tools
Git, LaTeX, Apache Airflow, Grafana, Jira, Microsoft Teams
Languages
Python, C, C++, Assembly
Platforms
Jupyter Notebook, Linux, Docker, Apache Kafka
Frameworks
Spark Structured Streaming, Hadoop, Yarn, Ray, Hydra
Storage
Apache Hive, Ceph
Other
Conda, Mathematical Analysis, Probability Theory, Linear Algebra, Machine Learning, Data Science, Deep Learning, Mathematics, Artificial Intelligence (AI), Statistical Methods, Computer Vision, Research, Analytics, Statistics, Slurm Workload Manager
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring