The problem
IFRS9 reporting at a major bank relied on large volumes of manual, spreadsheet-driven calculations. It was slow, hard to audit and easy to get wrong.
What I built
An in-house, automated IFRS9 system that pulls in financial data, runs the required calculations and produces the reports — end to end.
The backend combines Java services (Spring and Quarkus) with Python/Flask components for the data-heavy work, all containerised with Docker on top of Oracle.
Outcome
Manual processing time dropped by around 70%, with a consistent, repeatable process the finance team can trust.