Viktoria T. Data Science Engineer
Summary
Data Science engineer with over 3 years of practical commercial experience in Natural Language Processing (NLP), Computer Vision (CV), and Recommender Systems. Available skills in data analysis using machine learning approaches to satisfy business needs, problem-solving, and other tasks in this sphere. A person, focused on obtaining the best results, using all knowledge and skills. Friendly and ready to help the team complete tasks and solve certain problems.
Work Experience
Data Scientist (CV), Person Detection and Face Recognition
Duration: Duration: 1 year 1 months
Summary: The main goal was to classify people focusing on their actions and to recognize specific persons using their faces.
Responsibilities:
- Creating and engineering training data for a model.
- Implemented dataset preparation and model fine-tuning code for model evaluation.
- Evaluating different approaches to data preparation.
- Developing pipeline steps for the training model.
- Implementing new pre-trained models
Technologies: Python, Pandas, Pytorch, YOLO, RT-DETR
Data Scientist (NLP), PDF Text and Table Extraction
Summary: Extracting textual information from non-editable PDFs for quick collection and analysis. The task was related to recognizing and classifying text and tables in the picture using the "AWS Textract" service.
Technologies: Python, Matplotlib, Plotly, NumPy, Pandas, OpenCV, Regex, Spacy, AWS Textract
Duration: 4 months
Data Scientist, Online Retail Product Recommendations
Summary: Developed product recommendations system for online retail with analysis of historical user data. Developed a pipeline to generate popular items based on time, price, etc.
Responsibilities:
- Developed a recommendation system using implicit data
- Evaluated model offline and online (A/B tests) scores
- Developed model serving app and evaluated its performance
Technologies: Python, GCP, Tensorflow, Kubeflow
Duration: 6 months
Data Scientist (CV), Infrastructure Log Analysis
Summary: System for automated analysis of infrastructure logs which are row text data to discover the associated groups of resources that generate the logs based on extracted tags, ids, names, and other recognized entities.
Responsibilities:
- Using SQL-like databases for data extraction
- Extracting data from SQL-like databases based on specific task queries.
- Filtered and cleaned data to ensure accuracy and relevance.
- Merged multiple datasets into a single dataset to optimize extraction time and resources.
- Creating and filtering data using different patterns.
- Applying K-modes model for clustering words
- Preparing results taking into account the requirements of the
- customer
Technologies: Python, Matplotlib, Plotly, NumPy, Pandas, OpenCV, Regex, Spacy, AWS Textract
Duration: 8 months
Data Scientist (CV), Game Reward Management System
Summary: System for managing custom winnings during a game using user information such as user level, VIP rank, user wallet, etc. and generated probabilities for reward items.
Responsibilities:
- Implemented economy manager for receiving reward items.
- Created a simulation of the game using mock user data.
- Created pytests for verifying created algorithms.
Technologies: Python, Pandas, NumPy, SciPy, dataclasses
Duration: 4 months
Education
- Bachelor's degree in System analysis
2019 - 2023
Certification
- Data Science Camp Offline ML course at SmartInsight
2021 - Introduction to Data Science in Python
Coursera
2020 - Applied Machine Learning in Python
Coursera
2020 - Introduction to TensorFlow for Artificial Intelligence, Machine Learning, and Deep Learning
Coursera
2022 - Convolutional Neural Networks in TensorFlow
Coursera
2022 - Data Science Methodology
Coursera
2022 - Google Cloud Big Data and Machine Learning Fundamentals
Coursera
2023