kra_data_engineering

🇰🇪 KRA National Revenue & Trade Platform

Overview

This platform manages the automated ingestion and analysis of Kenya's national tax revenue, customs trade balances, and macro-economic indicators. It transforms data from PDF reports and structured files into actionable fiscal intelligence.

Architecture

graph TD
    PDF[Revenue Reports] -->|pdfplumber| IS[Ingestion Hub]
    Trade[TradeData.xlsx] -->|Pandas| IS
    IS -->|Load| P[PostgreSQL Raw]
    P -->|dbt Staging| Stg[Staging Layer]
    Stg -->|dbt Marts| Marts[Revenue & Trade Marts]
    Marts -->|Streamlit| Dash[KRA Executive Hub]
    Hub -.->|Orchestration| Airflow[Apache Airflow]

Loading

Data Sources

• KRA Revenue Reports: FY 2021-2025 audited performance.
• Customs & Trade: Integrated trade statistics and Rules of Origin disclosures.
• Economic Surveys: National GDP and industry growth rates (2026 Survey).

Tech Stack

• Orchestration: Apache Airflow
• Transformation: dbt Core (PostgreSQL)
• Database: PostgreSQL 15
• Visualization: Streamlit, Plotly
• Environment: Docker, Docker Compose

Folder Structure

kra_data_engineering/
├── dags/               # Fiscal & Trade ETL DAGs
├── dbt/                # Consolidated dbt project
├── ingestion/          # PDF parsing & Excel loading
├── dashboards/         # Visualization layer
├── tests/              # Data quality tests
├── docker-compose.yml  # Local stack definition
└── README.md

How to Run

1. Launch Stack:

   docker-compose up -d

2. Execute dbt:

   cd dbt
   dbt run
   dbt test

3. Access Dashboard: Open http://localhost:8506

Key Metrics / Outputs

• YoY Revenue Growth: Performance by primary tax heads.
• Trade Balance: Import/Export volume by HS-code commodity.
• Tax-to-GDP Ratio: National tax efficiency modeling.