Back to Projects

Client project

IFRS 17 Desktop Applications: GMM, VFA and PAA

Three PySide6 desktop applications (PAA, GMM, and VFA) that drive macro-enabled Excel models for IFRS 17. I built the interface and integrated it with the model components.

  • PySide6
  • xlwings
  • IFRS 17
  • PyInstaller
  • Pandas
Confidential work: no public demo or source code

Overview

Three Windows desktop applications for IFRS 17 measurement, covering the premium allocation approach (PAA) for non-life business and the general measurement model (GMM) and variable fee approach (VFA) for life business. A further variant of the VFA application is released separately with its own model. Each application runs an Excel-based actuarial model end to end: inputs, data checks, model execution, consolidation, and output files.

The problem

The actuarial models live in large, macro-enabled Excel workbooks. Running them by hand across entities and portfolios is slow and error-prone, and actuaries needed a guided application that validates inputs before long-running calculations start.

My role

I built the desktop interface and integrated it with the backend and model components, working alongside another developer and the actuarial team, who own the calculations and validation. I also adapted the original non-life application into the life insurance versions for GMM and VFA. The VFA variant reuses the VFA application I developed; for that release I integrated its updated backend model and updated its title and metadata.

How it works

  1. Upload the Excel inputs.
  2. Data check: inputs are validated before any long processing starts.
  3. Run the model: Python drives the Excel workbooks and their macros.
  4. Consolidate results across entities and portfolios, then export the output files.

Engineering highlights

  • Shared architecture, independent releases: the applications share a UI and application structure, while each keeps its own model logic and is released separately, which made the VFA variant a model swap rather than a new build.
  • Excel automation: xlwings drives workbooks and macros; openpyxl handles direct workbook edits.
  • Multi-page desktop UI with shared state across pages.
  • Packaged executables: separate path handling for development and packaged builds, and a release checklist that tests the packaged app end to end with representative inputs.

Stack: Python, PySide6, xlwings, openpyxl, Pandas, NumPy, PyInstaller.

Outcome

A guided desktop workflow for IFRS 17 models across non-life and life business, so actuaries can run validated model cycles with less manual handling of workbooks.

Client data, model workbooks, and the actuarial methodology are confidential and are not shown here.