praful yelekar

At Highspring India, I design and automate end-to-end data pipelines using Snowflake, PySpark, Apache Airflow, and AWS EMR. My work has reduced manual monitoring by 15+ hours per month while improving pipeline reliability.

I build data validation and quality checks for null handling, deduplication, and schema validation, improving downstream BI and analytics accuracy to approximately 95%. I also work with cross-functional teams to design Star and Snowflake data models for reporting and decision-making.

I've built automotive data pipelines that ingest FTP data, standardize it into Parquet, stage it in Amazon S3, and load validated datasets into Snowflake for reporting dashboards.

For crypto market analytics, I developed a near real-time ETL pipeline running every 15 minutes with S3, Glue, PySpark, Redshift, Lambda, and EventBridge. I optimize Spark transformations, SQL queries, partitioning, and warehouse models to support reliable, scalable reporting.