At CGI, I integrated heterogeneous data sources into a unified ETL pipeline for a large energy client, reducing onboarding time for new sources by 40%. I also built real-time ingestion workflows processing over 500K records daily across Bronze and Silver layers.
At Fondation des Oeuvres Sociales du MEF, I built a document-processing ETL pipeline for AI chatbot data ingestion. I developed workflows to prepare corpora for NLP and optimized a relational repository, improving query response time by 35%.
On my Cloud Data Warehouse project, I developed an end-to-end pipeline with PostgreSQL and GCP, adding 15+ data quality tests and dashboards for film performance and customer trends. I also designed a real-time Reddit fake news detection pipeline with batch and streaming ETL.
