Ovchinnikov Grigoriy
Verified Expert in Engineering
Data Scientist and Developer
Tbilisi, Georgia
Toptal member since July 14, 2022
Ovchinnikov is a data scientist with over three years of data analysis and machine learning experience. He has a strong knowledge of classical ML algorithms, statistics, and proficiency while working with ML frameworks and databases for solving problems. Ovchinnikov is also well-versed in handling big data for modeling and data analysis.
Portfolio
Experience
- Data Science - 5 years
- Python - 5 years
- Unsupervised Learning - 4 years
- Machine Learning - 4 years
- Supervised Machine Learning - 4 years
- PySpark - 3 years
- Advertising Technology (Adtech) - 2 years
Availability
Preferred Environment
PyCharm, Jupyter Notebook, GitLab, Ubuntu
The most amazing...
...projects I've worked on include solutions for several problems requiring thinking outside the box, which significantly positively impacted product development.
Work Experience
Middle Data Scientist
Platforma
- Implemented state-of-the-art methods for a look-alike modeling task with positive unlabeled (PU) learning using PySpark and PyTorch with a significant uplift in offline and online metrics.
- Built an ML model for labeling clickstream data with target device information using a weekly supervision paradigm which increased accuracy and lower inference time on telecom data in more than hundreds of million devices using PySpark and Hadoop.
- Suggested and implemented an NLP-inspired approach solution for website (host) representation learning based on clickstream data, giving several advantages over the previous approach.
- Developed a method for generating user interest vectors based on clickstream data embeddings.
Junior Data Scientist
S7 Airlines
- Built a decision support system based on clustering and semi-supervised multiclass classification using a traditional "tabular" machine learning stack.
- Created a pipeline for scraping external open data with Python using mainly Scrapy, Airflow, and PostgreSQL.
- Worked on a solution for predictive maintenance tasks on remaining useful life estimation.
- Contributed to a big data tech stack by performing an exploratory data analysis and building MVP projects.
- Built supervised ML regression models for predicting the number of sold tickets for flights.
Experience
ID R&D Facial Antispoofing Challenge
As a result, I built a solution to achieve a ROC AUC score greater than 0.99 based on siamese neural networks. My approach was amongst the top 6 best solutions.
Catch Blogger
Pipeline stack: Python, Puppeteer, Google Cloud Platform (GCP) including Cloud Functions, Cloud Scheduler, and Cloud Storage.
Education
Bachelor's Degree in Computer Science
Volgograd State Technical University - Volgograd, Russia
Certifications
Drawing Inferences from Data
Moscow Institute of Physics and Technology | Yandex | E-learning Development Fund
Skills
Libraries/APIs
Scikit-learn, NumPy, Pandas, PySpark, PyTorch
Tools
Apache Airflow, PyCharm, Git, GitLab
Languages
Python, SQL
Platforms
Jupyter Notebook, Docker, Ubuntu
Frameworks
Hadoop, Scrapy
Storage
PostgreSQL, Amazon S3 (AWS S3), MongoDB, Google Cloud
Paradigms
Agile Software Development, ETL
Other
Machine Learning, Gradient Boosting, Data Science, Unsupervised Learning, Supervised Machine Learning, Deep Neural Networks (DNNs), Positive-unlabeled (PU) Learning, Advertising Technology (Adtech), Interpretation, Web Scraping, Programming, Informatics, Applied Mathematics, Analytics, Statistics, Data Analysis, Hypothesis Testing, Personalization, Deep Learning, Computer Vision Algorithms, Natural Language Processing (NLP), Generative Pre-trained Transformers (GPT)
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