← All work

02 / Machine Learning / Real-Time Systems

Public · Active development

Real-Time Fraud Detection Platform

A production-style fraud platform that turns payment events into calibrated risk scores, policy decisions, alerts, drift reports, and analyst-facing operational views.

Built an end-to-end system that separates model risk from business policy while connecting event streaming, point-in-time features, explainability, monitoring, and review workflows.

TRANSACTION STREAMRISK SIGNAL

01 / PROBLEM

What the system is trying to solve.

A useful fraud system needs more than a classifier. It has to score events reliably, preserve event-time correctness, expose why a decision happened, and continue behaving predictably as data changes.

02 / SYSTEM

How it works.

  1. 01

    Ingest payment events through a Kafka-compatible Redpanda stream.

  2. 02

    Build leakage-safe chronological features and score events through a FastAPI inference service.

  3. 03

    Apply APPROVE, REVIEW, and BLOCK policy thresholds separately from model probabilities.

  4. 04

    Persist decisions through idempotent consumers and a transactional outbox pattern.

  5. 05

    Track delayed-label performance, SHAP explanations, drift statistics, and champion/challenger retraining signals.

03 / ENGINEERING

Decisions that shaped the project.

01

Leakage-safe temporal evaluation and point-in-time feature design.

02

XGBoost model experimentation with MLflow tracking and Optuna tuning.

03

SHAP-backed explanation and alerting workflows.

04

PSI, KS, and JS drift monitoring plus delayed-label metrics.

05

Next.js command center with operational views and Playwright end-to-end coverage.

04 / EVIDENCE

A few concrete signals.

0.9266

Precision

0.3108

Recall

0.4654

F1

0.3363

PR-AUC
  • Python
  • FastAPI
  • Redpanda
  • PostgreSQL
  • XGBoost
  • MLflow