Client project
IFRS 17 Cashflow Engine
A full-stack IFRS 17 cashflow and valuation engine for macro-enabled Excel inputs that cut run time from about a week to a day.
Overview
A cashflow and valuation engine for insurance reporting under IFRS 17. Users upload gross business workbooks and, optionally, reinsurance workbooks; the engine projects cashflows, calculates the liability for incurred claims (LIC) and the liability for remaining coverage (LRC), reconciles the outputs, and produces the Excel workbooks used for reporting.
The problem
The inputs are large, macro-enabled Excel workbooks, and the calculations previously took around a week to run and assemble. Every run also needed careful checks that outputs reconciled with inputs.
My role
I was responsible for the frontend and backend, including integrating the actuarial modelling code with the application. The actuarial team owns the calculation methodology and validation; I built the application around it and contributed to the performance work.
How it works
- Upload and validate. The app reads the required sheets from macro-enabled workbooks and validates them before processing.
- Set parameters. Users choose the valuation and projection settings for the run.
- Run. The engine performs cashflow, LRC, reinsurance, and output-reconciliation logic, with live progress shown in the UI.
- Review and download. Output workbooks and validation summaries, individually or as a single archive.
Engineering highlights
- Layered design: routes, a separate calculation service layer, and shared configuration.
- Config-driven sheet loading: workbook structures are declared once instead of hard-coded across the code.
- Reconciliation built in: every run produces validation and consistency checks alongside the results.
- Isolated runs: each session writes to its own output folder.
- Flexible delivery: runs as a web server or as a desktop window.
Stack: Python, Flask, Pandas, NumPy, openpyxl, XlsxWriter, Waitress, PyWebview.
Outcome
Processing time dropped from about a week to about a day, and the platform has continued to evolve through several major versions.
Client data, templates, and the actuarial methodology are confidential and are not shown here.