BSc Data Science student at KCA University | Expected graduation: September 2027
Focused on Data Science, Machine Learning, Applied AI, and Generative AI. I build practical projects in predictive analytics, explainable machine learning, and natural language processing (NLP).
- Programming and data: Python, SQL, SQLite, pandas, NumPy, Excel
- Machine learning and explainability: scikit-learn, Logistic Regression, Random Forest, GridSearchCV, SHAP; predictive modeling, feature engineering, cross-validation, model evaluation
- Data visualization: Matplotlib, Seaborn, Plotly
- Applied AI: Rule-based NLP; retrieval-augmented generation (RAG), vector search, and Generative AI through ongoing academic work
- Applications and development tools: Streamlit, Jupyter Notebook, VS Code, Git, GitHub, pytest
Explainable financial transaction fraud detection using Random Forest and SHAP, built with Python and Streamlit. Supports single-transaction and batch predictions, with 47 passing automated tests.
Tech: Python, pandas, NumPy, scikit-learn, Random Forest, SHAP, Streamlit, pytest
Final held-out evaluation:
| Metric | Result |
|---|---|
| Precision | 0.945946 |
| Recall | 0.736842 |
| F1 | 0.828402 |
| PR-AUC | 0.809520 |
These metrics come from the final historical holdout evaluation. Synthetic demonstration transactions are used only to demonstrate the interface and are excluded from these evaluation results.
Customer churn classification developed during the Lloyds Banking Group Data Science Job Simulation. Covers data preprocessing, exploratory analysis, feature engineering, Logistic Regression, hyperparameter tuning with GridSearchCV, cross-validation, and model evaluation.
Tech: Python, pandas, scikit-learn, Matplotlib, Seaborn
Rule-based Python chatbot developed during the BCG GenAI Job Simulation. Answers predefined financial questions using structured financial data and pandas analysis.
Tech: Python, pandas, rule-based NLP, financial analysis
Python and SQLite inventory management system supporting authentication, product management, stock tracking, sales recording, reporting, and persistent database storage.
Tech: Python, SQLite, VS Code
KCA University Intelligent Academic Advisor — Developing an academic advisor using RAG and Generative AI as an ongoing university project.
The following experiences are job simulations, not employment roles.
- Customer churn analysis and predictive modeling using Logistic Regression and GridSearchCV.
- Data preprocessing, feature engineering, model evaluation, and feature analysis.
- Customer review data analysis and data visualization.
- Predictive modeling of customer buying behaviour.
- Financial analysis of Microsoft, Tesla, and Apple 10-K data using Python and pandas.
- Development and testing of a rule-based financial chatbot.
- Machine learning, Generative AI, RAG, and NLP.
- Advanced Python and SQL.