IMBIshimwe Mugisha Benjamin

Case study · 04 / 10

National regulation platform

A secure, always on backend for a national regulator, with a columnar analytics pipeline.

Organisation
Loxotech Ltd
Role
Software engineer
When
Apr 2025 to Mar 2026
System view of National regulation platform: the architecture, drawn without private data
System view. The real screens are private, this is how the system fits together.

Context

A regulator needs one place for records that must be correct, auditable and available under unpredictable public load. Screens stay private, so this page shows the system, not the UI.

What I built

  1. 01The secure backend architecture and REST API layers, built for regulatory compliance.
  2. 02Data pipelines in Python with Polars and Parquet behind a dynamic analytics platform.
  3. 03CI/CD and high availability Kubernetes environments, co managed with the team.

Outcome

  • 30% faster query processing in the analytics pipeline.
  • 99.9% uptime under variable public sector load.

Tools

  • Python
  • Polars
  • Parquet
  • PostgreSQL
  • Kubernetes
  • GitHub Actions