Built an internal agentic tool on Claude Code (custom subagents + tools) to map dependencies between SQL Server stored procedures and data warehouse tables, exposing undocumented data flows and unblocking a large-scale migration project. Developed an internal RAG agent on Langflow over a Confluence knowledge base to serve data product documentation to business users — rewrote the chunking logic with additional split patterns and iterated on retrieval strategy, reducing hallucinations and improving answer relevance. Designed a CLI for cross-validation between Hudi tables, DuckDB, and other sources — orchestrated through Claude Code subagents and commands — that automated data-consistency checks across storage layers and caught discrepancies before production, preventing downstream rework. Built internal developer tooling with Claude Code (custom skills, subagents, slash commands), including a testing-acceleration agent that eliminated manual file uploads and throwaway PRs from the dev loop. Designed and maintained data pipelines using Python, Airflow, AWS, and Spark. Improved pipeline performance and reduced failures by optimizing orchestration and data integration.

Built and maintained ETL pipelines using Python, Airflow, AWS, and Spark. Implemented Databricks to improve data processing performance and analytics capabilities. Designed large-scale ingestion pipelines using Kafka and PySpark, processing billions of daily events from marketplace activity. Acted as the only Data Engineer, owning AWS infrastructure, ingestion pipelines, and orchestration workflows, ensuring platform reliability and scalability. Led the migration from Airflow to Databricks, improving scalability and enabling more efficient data processing.

Built pipelines integrating multiple data sources across cloud environments. Supported financial data workflows and reporting. Led the migration of a legacy SQL Server data warehouse to a Snowflake-based lakehouse, improving performance and scalability. Reverse-engineered business logic from stored procedures to ensure accurate data transformation. Maintained data operations during a team transition, acting as the primary Data Engineer for approximately 1.5 months.

Built ETL pipelines and data workflows using Python, Airflow, Alteryx, and AWS. Supported analytics and reporting for business teams. Re-architected memory-intensive pipelines into in-database processing (Redshift + Airflow), eliminating recurring failures, reducing approximately 10+ hours/day of manual effort, and improving data delivery reliability.

Developed ETL workflows to process and model legal data for analytics.

Built dashboards and automated reporting solutions using Tableau, Power BI, SQL, and Excel.