At McKinsey & Company, I replaced an Excel price simulation tool with a Python, Streamlit, and PostgreSQL application, cutting a day-long update process to about five minutes. Clients can upload files to update calculations and run quality checks.
I also brought fragmented promotion data into a consistent analysis for a client working with multiple U.S. retailers, and proposed a data stack for ongoing promotion tracking. I helped client teams implement revenue-growth analyses and adapt the code to run analyses end to end.
For incoming client files, I built a Dagster pipeline with bronze, silver, and gold layers and automated data quality checks. I also built Databricks pipelines and jobs with Spark and PySpark; earlier, at BAC Credomatic Latam, I analyzed transactions and developed regional fraud detection rules.
