Work / Public sector
Data and economics · Public sector · 2026
Sisbén IV data pipeline
The first Power BI dashboard of Antioquia's Planning Department and the Python pipeline behind it, built on ~25M records of Sisbén, the survey Colombia uses to target social programs.
- Status
- Covers all 125 municipalities
- Role
- Data & Economics Intern
- Period
- Feb – Jun 2026
- Stack
- Python · DuckDB · PyArrow · Power BI · DAX · TMDL
Abstract
Sisbén is the survey Colombia uses to decide who can receive social programs, so its numbers have to be right before anyone reads a chart. I built the path from six raw annual snapshots to a dashboard by municipality: a Python pipeline that cleans and audits the data, and a Power BI model the department’s Technical Director used with all 125 municipalities of Antioquia.
What I built
- The first Power BI dashboard of Antioquia's Planning Department, with Sisbén data by municipality (2021–2026), used by the Technical Director with all 125 municipalities.
- A batch ETL in Python with DuckDB and PyArrow over ~25M records across six annual snapshots (2021–2026).
- 24 derived columns moved into the batch ETL, plus automated data-quality audits.
- A Power BI semantic model in TMDL and DAX, with 19 tables.
| What | Count |
|---|---|
| Records | ~25M |
| Annual snapshots | 6 (2021–2026) |
| Derived columns | 24 |
| Tables in the semantic model | 19 |
| Municipalities in the dashboard | 125 |