I build data pipelines and data processing systems with an emphasis on rule-based behavior and explicit data contracts to produce analytics-ready outputs.

• Eurostat Trade Pipeline – Batch Ingestion and Transformation

  • Downloads and extracts historical monthly COMEXT bulk files into an immutable raw layer
  • Transforms raw .dat files into clean silver Parquet via DuckDB SQL
  • Containerized with Docker Compose (separate ingestion, transform, and pipeline services)
  • Idempotent re-runs, explicit data contracts, and CI with Ruff + ShellCheck
    → https://github.com/bravojuandb/eurostat-trade-pipeline

• Navarra Economic Establishments – Batch Data Pipeline

  • Rule driven batch pipeline for cleaning and validating public administrative data
  • Explicit handling of schema, null semantics, and identifier fields
  • Produces analytics-ready Parquet output
  • No database load
    → https://github.com/bravojuandb/navarra-data-batch-pipeline

📧 bravojuandb@protonmail.com

📍 Navarra, Spain · 🌍 Open to remote opportunities