At LTIMindtree, I've migrated 30+ legacy Snowflake stored procedures into PySpark transformation logic across bronze, silver, and gold lakehouse layers. I reduced manual conversion effort by 40–60% by refining watsonx.ai Granite prompts for automated Spark code generation while preserving data accuracy.
I've delivered audit-ready pipelines through Git and CI/CD, applied column-level data masking controls, reconciled migrated datasets to achieve zero unresolved discrepancies at sign-off, and mentored junior engineers on Snowflake-to-lakehouse migration practices.
