CHILEAN AUTOMOTIVE ECOSYSTEM

Data Pipelines - Web app + ETLs to extract and clean data from the automotive sector

Secciónes

Stacks

🐍 Python • ⚛️ React • 🤖 Selenium • ✨ Gemini API • 🐼 Pandas • 🔌 API REST

Link Repositorio Github

Problema

Data on the Chilean automotive market is fragmented: sales figures are in ANAC PDFs, accidents are in public APIs, and licenses are in INE Access databases. Each source has different formats, changing structures, and requires manual cleaning. An analyst can waste days just preparing data before generating value.

Solución

VentasExtract

Web app that extracts sales tables from ANAC PDFs using AI (Gemini) Unstructured PDF → Clean CSV in minutes

ETL Accidents

Pipeline that consumes road safety API and generates historical series 2018-2024 Data ready for temporal and regional analysis

Integration of INE databases (licenses + permits) with category approval Unified dataset of the Chilean vehicle fleet

ETL Licenses