Open to opportunities

Elizaveta Soldatenkova

@elizavetasoldatenkov

Message

Data Scientist focused on predictive modeling and ML-driven automation.

Russian Federation

Message

What I'm looking for

I’m looking for a Data Scientist role where I can build predictive models, automate analytical workflows, and apply behavioral NLP—working in a team that supports rigorous experimentation and measurable business impact.

I’m a Data Scientist with 1 year of experience in machine learning, specializing in predictive modeling and data analysis. I’ve delivered hands-on solutions across industry and research, including an internship at Gazprom Neft Science & Technology Center and academic work in behavioral NLP.

At Gazprom Neft, I built ML solutions that automated analytical workflows and improved forecasting accuracy by 12–54%. I developed an NPV forecasting tool that streamlined evaluation for 3,750 investment projects and improved model performance from R² = 0.84 to R² = 0.96 using XGBoost and LightGBM with hyperparameter optimization. I also created a portfolio optimization tool that reduced portfolio analysis time from 3 days to less than 1 minute, incorporating business constraints and generating analytical reports via interactive dashboards.

Experience

Work history, roles, and key accomplishments

Data Scientist

Behavioral AI Research Team

Sep 2025 - Jun 2026 (9 months)

Researched and developed a deep learning model for predicting user behavior in online discussions using behavioral NLP. Reduced prediction error (MSE) by 54% (0.0070 to 0.0032) using Optuna-based hyperparameter optimization and an LSTM model with an Attention mechanism.

Deep Learning Attention Mechanisms Feature Engineering Optuna Sequence Modeling data preprocessing

Data Scientist Intern

Gazprom Neft Scientific and Technical Centre

Jun 2025 - Aug 2025 (2 months)

Developed analytical tools to automate oil-and-gas investment project evaluation, including an NPV forecasting tool and an investment portfolio optimization tool. Improved NPV forecasting accuracy by 12%, increased model R² from 0.84 to 0.96, and reduced portfolio analysis time from 3 days to under 1 minute.

XGBoost LightGBM RandomizedSearchCV Chi Square Test data preprocessing Dashboard Development Portfolio Optimization

Education

Degrees, certifications, and relevant coursework

Kazan Federal University

Bachelor of Information Systems and Technologies, Information Systems and Technologies

2023 - 2027

Grade: 4.95/5.0

Activities and societies: Relevant coursework: Data Science, Machine Learning, Mathematical Statistics, Probability Theory, Linear Algebra, Algorithms and Data Structures, Deep Learning Basics.

Bachelor's program in Information Systems and Technologies with a GPA of 4.95/5.0 and coursework covering core data science and machine learning topics.