I build at the intersection of data, risk, and real-world decision-making.
My background combines Computer Science and Crime Science with Data Science (UCL BSc and MSc), and I’m currently focused on using SQL, Python, statistics, and machine learning to understand and solve high-stakes problems in areas such as:
- Financial crime / AML transaction monitoring
- Fraud and payments risk
- Credit risk and risk analytics
- Explainable AI for regulated environments
Right now, I’m building a portfolio around risk-oriented quantitative analysis, with projects that move across three layers:
- Detect risk — AML / transaction monitoring / fraud analytics
- Quantify risk — credit risk / scoring / model evaluation
- Use risk and signal to make decisions — quant-style research and strategy thinking
- Strengthening SQL as a core analytical tool
- Building transaction-level analytics projects
- Comparing rule-based and ML-based risk scoring approaches
- Learning how to turn messy real-world data into clear business decisions
- Studying how explainability, governance, and regulation shape data work in finance
Languages & tools
SQL Python Pandas NumPy scikit-learn XGBoost SHAP Jupyter Excel
Core themes
Data Analysis Risk Analytics AML Transaction Monitoring Fraud Detection Credit Risk Machine Learning Explainable AI
This profile is where I document my progress from:
- structured query and data analysis
- to risk scoring and model evaluation
- to more advanced quantitative and decision-focused work
My repositories are designed to be more than coding exercises — I try to frame them around real business problems, realistic constraints, and clear analytical reasoning.
I’m especially interested in how data can make complex systems more legible.
For me, statistics and analytics are not just technical tools — they are ways to:
- understand uncertainty
- detect meaningful signals
- support better decisions
- and connect technical work to real institutional and social impact
- Advanced SQL for analytical workflows
- Transaction monitoring system logic
- Credit risk modelling concepts
- Quantitative reasoning under uncertainty
- Model comparison, validation, and interpretability
- Conversations around financial crime analytics, risk analytics, and quantitative portfolio projects
- Learning from strong examples in regulated AI, fraud/risk modelling, and financial data systems
- Collaborating on projects where data meets real-world systems
Building a long-term portfolio across risk detection, risk measurement, and decision-making under uncertainty.