Worked on the apache/auron repository to address a critical issue in the OrcScan component, focusing on improving data ingestion reliability for big data workflows. Refactored the schema mapping logic to ensure that missing data columns are correctly handled and that projection mapping remains robust, which eliminated data inconsistencies during reads. This technical approach involved careful validation to account for all required columns, reducing the risk of schema-related failures in production environments. Leveraged expertise in Python, Rust, and distributed systems to deliver a targeted bug fix that enhanced downstream analytics by ensuring higher data quality and more dependable data engineering pipelines.