I will build a fraud detection and anomaly detection system

Fraud costs businesses billions every year. I build machine learning fraud detection and anomaly detection systems that flag suspicious activity in real time before it becomes a loss.

I specialise in imbalanced classification problems (fraud is rare by nature), using XGBoost, Random Forest, and isolation forests with SMOTE oversampling to build models that don't miss fraudulent transactions. The result is a system your team can trust and act on.

What you get:

• Exploratory data analysis on your transaction data
• Trained fraud or anomaly detection model
• Threshold tuning for your precision vs recall tradeoff
• Interactive dashboard for monitoring flagged transactions
• Model evaluation report (F1, recall, precision, AUC)
• Clean, documented Python codebase

Who this is for: fintech companies, e-commerce platforms, payment processors, and any business handling transaction data that needs a smarter way to catch bad actors.