Felipe Marra

Created a dynamic Databricks website/app to document approximately 6,000 legacy jobs and pipelines by engineering an ETL pipeline using multiple LLM agents and the Databricks Genie within 3 months, reducing maintenance time and simplifying the debugging process from days to hours. Mapped and cataloged over 55,000 data tables into the Data Lake Unity Catalog by developing an AI-powered data pipelin

Tested business hypotheses using Python, PySpark, and SQL to optimize complex scalable ETL/ELT pipelines. Evaluated and disproved an inefficient supply chain hypothesis by conducting a data-driven study on product distribution, preventing a projected increase in deadstock and unsellable goods.

Optimized product distribution and phygital stock management strategies by leveraging advanced data models. Developed data pipelines for retail item characterization and performance tracking to support business decision-making.

Coordinated the development of an e-commerce predictive routing model, reducing shipping costs by 15%. Co-developed and maintained scalable ETL pipelines utilizing Azure Data Factory (ADF) and Databricks.

Automated manual routines using Python and Selenium web scraping, cutting form-processing time by 80%. Engineered the core vehicle financing engine and updated credit indicators via SQL routines.

Architected real-time stockout management algorithms for Black Friday, successfully achieving a 0% stockout rate. Developed complex data visualization projects and dashboards using Tableau, BigQuery, and Oracle SQL.

Pursued an MBA in Machine Learning, focusing on advanced machine learning techniques and applications.

Completed a Bachelor's Degree in Data Science, covering data analysis, statistics, and machine learning fundamentals.

Earned an Engineer's Degree in Materials Engineering, with a focus on materials science and engineering principles.

Studied Manufacturing Engineering and Materials Science as part of an exchange program.