Machine Learning Engineer | Data Scientist | Innovator in Fraud Detection & AI-Powered Solutions
A Machine Learning Engineer and Data Scientist with a strong focus on building real-world, scalable solutions in fraud detection and predictive modeling. My work blends deep analytical thinking with a practical mindset, always aiming to make data science tools more accessible, efficient, and impactful.
I’ve developed end-to-end machine learning pipelines, worked with unbalanced and complex datasets, and designed solutions that improve decision-making and risk management. I enjoy working across the full data lifecycle — from data exploration and feature engineering to modeling, evaluation, and communication of insights.
My technical stack includes Python, SQL, and tools like scikit-learn, XGBoost, Kafka, and Tableau. I care deeply about clean, interpretable models and the user experience around data products. Whether working independently or in a team, I’m driven by curiosity, clarity, and the desire to make data truly useful.
Supervised learning, model evaluation, class imbalance handling
Python, SQL, Git – clean, readable code focused on clarity, reproducibility, and collaboration
Insightful dashboards and reports for technical and non-technical audiences (Tableau, seaborn)
EDA, data cleaning, feature creation, hypothesis generation using pandas, NumPy, matplotlib
Python | EDA | Feature Engineering
Built a machine learning pipeline to detect fraudulent transactions in a synthetic FPS dataset. ▪ Performed EDA and created behavioral and ratio-based features to capture suspicious activity. ▪ Handled strong class imbalance using SMOTETomek and class weighting. ▪ Trained and evaluated multiple models (LogReg, Random Forest, XGBoost); Random Forest achieved the best F1-score. ▪ Focused on reproducibility, clean code, and clear documentation.
Exploring Financial Fraud - Project Summary
Python | Streamlit | API Integration | Data Visualization
An interactive web app that helps users identify the best countries to relocate to, based on their personal and professional priorities. It combines World Bank and HDI data, normalizes indicators like GDP, inequality, education or inflation, and calculates a relocation score. Users can customize weights, compare countries in radar plots, and explore forecasts using machine learning.
Data-Driven Relocation – Project Summary
