← В ленту
Регистрация: 20.03.2025

Артур Межлумян

Специализация: Data Scientist / Machine Learning Engineer
— Machine Learning Engineer with a strong background in predictive modeling, deep learning, and data-driven decision-making. — Experienced in developing and deploying ML solutions for tabular, time-series, and NLP data. — Passionate about optimizing models for scalability and real-world impact. — Machine Learning & Deep Learning – exploring new architectures and improving models. — Natural Language Processing (NLP) – working with text, chatbots, and LLMs. — Time-Series Analysis & Forecasting – forecasting data (e.g., stock market, sports). — MLOps & Model Deployment – automating ML processes, CI/CD, monitoring. — Data Science for Finance & Sports Analytics – applying ML in finance and sports.
— Machine Learning Engineer with a strong background in predictive modeling, deep learning, and data-driven decision-making. — Experienced in developing and deploying ML solutions for tabular, time-series, and NLP data. — Passionate about optimizing models for scalability and real-world impact. — Machine Learning & Deep Learning – exploring new architectures and improving models. — Natural Language Processing (NLP) – working with text, chatbots, and LLMs. — Time-Series Analysis & Forecasting – forecasting data (e.g., stock market, sports). — MLOps & Model Deployment – automating ML processes, CI/CD, monitoring. — Data Science for Finance & Sports Analytics – applying ML in finance and sports.

Скиллы

Python
Pandas
NumPy
Seaborn
Scikit-learn
TensorFlow
MySQL
NLP
Keras
OOP

Опыт работы

Data Scientist / Machine Learning Engineer
с 2025 - По настоящий момент |NDA
CI/CD, Python, Pandas, Seaborn, Scikit-learn, MySQL
● Developed and deployed ML models for the banking sector (lending and refinancing). ● Optimized and tested models using modern frameworks. ● Built pipelines for handling large datasets, including data collection, cleaning, and analysis. ● Deployed models into production using CI/CD tools. Project: Kaggle & lending and refinancing. ● Real Estate Price Prediction (Kaggle). ● Used machine learning methods (XGBoost, LightGBM) to predict real estate prices. ● Achieved high accuracy with less than 5% error in price predictions. ● NCAA Tournament Outcome Prediction (Kaggle). ● Developed models to predict NCAA tournament results, improving accuracy by 10% over base models. ● Credit Default Prediction (Kaggle). ● Developed models for credit risk assessment using class balancing and feature engineering techniques.

Образование

Data Science Program
2024 - 2025
Smart Code

Языки

РусскийРодной