I built a reusable, config-driven ingestion pipeline in Azure Data Factory that processes files across business folders without hardcoded names, paths, or file types.
For the pipeline, I used a master-and-worker design with parameterized execution, file validation, empty-file detection, and a success-gated copy, archive, and delete sequence.
I also built a full and incremental load pipeline for Azure SQL Database, using watermark tables to load changed records and keep target tables consistent without reprocessing unchanged data.
At Innodata, I processed and validated structured and unstructured datasets for AI applications, ran SQL data quality checks, and automated repetitive processing with Python. At SCS Tech, I investigated production issues with SQL and application logs and troubleshot database issues.

