creditanalytics

This project is a machine learning-based web application that predicts the likelihood of credit card approval for a customer based on their personal and financial details. It leverages a Random Forest Classifier trained on historical credit and application records.

Interactive DataQuest 2026 credit-risk analytics platform combining EDA, statistical research, interpretable logistic regression, scorecard concepts, and business decision support. The project focuses on understanding borrower risk, identifying meaningful patterns, evaluating model performance, and translating predictions into practical lending.

R-based financial risk analytics engine and Shiny dashboard — debtor risk scoring, collections notifications, a trained repayment-prediction model, and an LLM-powered natural language chat layer, built on real Kaggle datasets.

An interactive exploratory data analysis (EDA) and advanced visualization project in R investigating 30,000 credit card default histories in Taiwan. Uses PCA feature screening, LOESS smoothing curves, and color-blind friendly Plotly dashboards to isolate key financial risk drivers across distinct credit limit tiers.

A lending product's risk engine, phase by phase — from a cold-start expert scorecard to a data-driven model trained on real repayment outcomes.

PostgreSQL analytics on 1.75M real New York mortgage applications, 2022 to 2025. Star schema, quality assertions, performance tuning, generated data dictionary and a browser-runnable SQL database/playground that needs no installation.

Projeto de ETL e análise de dados das operações de crédito referentes a 2025, disponibilizados pelo Banco Central do Brasil por meio do Sistema de Informações de Créditos (SCR).

Development of PANorama, a credit limit optimization platform built with Banco PAN. Using Linear Programming and a custom Simplex solver validated with OR-Tools, the solution optimizes credit allocation at scale. Built with FastAPI, React, PostgreSQL, Redis, Celery, and Docker, it was validated with over 2.8M customers.

Simple Credit Risk/Fundamentals screener

SQL (SQLite) project focused on credit lifecycle management and operational risk control. Features end-to-end schema definition, data population, and analytics queries for funnel conversion rates and default/late payment identification.

Early-warning dashboard for credit-cycle stress. Tracks Fed lending standards, high-yield issuance, CCC spreads and HYG/LQD into a traffic-light HTML report. Python, no API keys.

Analysis case and working prototype for auditing a fleet credit book after contracts are booked. 14 SQL rules (DuckDB) and a Python scoring layer flag mispriced, split and double-financed deals, tested against planted problems on fully invented data. Ends in a pilot proposal.

Credit Analyst Job Simulation by Standard Chartered through Forage!