Wasi Mahin

The Problem

Manual PDF-to-ERP data entry consumed significant staff hours per batch. No API existed. No ERP credentials available. The only path was a human typing into Business Central, field by field.

The Approach

Zero-API desktop RPA: parses vendor invoice PDFs with pdfplumber, extracts structured data, then drives the Business Central UI directly via pyautogui. No credentials, no integration, just results.

The Problem

Journal entry processing required manually reading DocuSign PDFs and CashNet Excel files, then re-entering every debit/credit line into Business Central. Slow, error-prone, and done repeatedly.

The Approach

Automated end-to-end: reads PDF and Excel sources, structures debit/credit lines, then enters complete journal entries into Business Central via hotkey automation. No manual re-keying.

The Goal

Build a real-world BI problem from scratch: combine disparate datasets, design a data model, and create an interactive dashboard that tells a narrative story rather than just displaying numbers.

The Result

Treats bounty as a proxy for value and risk, analyzing distribution across crews, arcs, and regions. Unified Kaggle datasets into a single model; Chart.js visualizations let users explore patterns interactively.

The Question

Is economic mobility still achievable for the average household? The team went beyond surface-level income data to interrogate purchasing power, regional inequality, and structural cost-of-living shifts.

The Analysis

End-to-end economic research: income growth vs. household purchasing power, cost-of-living adjustments, regional wealth inequality, visualized in Tableau with a clear narrative arc.