GX Cloud & OSS Library
Technical Design, Performance Optimization, AI Agent, Open Source
- Authored technical designs and ADRs for major features, then led implementation across the full stack (React SPA, Python microservices, GraphQL APIs, open-source core library). Added SQL Server and Microsoft Fabric as fully supported data sources end-to-end, opening the product to a new customer segment.
- Designed and shipped filtering and drill-down for a data quality observability dashboard, letting users scope all metrics by datasource, asset, or column. Wrote the technical design, built the backend query engine optimized for PostgreSQL row-level security, and implemented the React UI.
- Diagnosed and resolved production database performance issues, reducing query latency from 20–46s timeouts to under 400ms by rewriting SQL to defeat PostgreSQL generic-plan caching regressions under row-level security. Optimized dashboard queries by up to 1,590x through join-order and indexing changes.
- Broadened an AI agent for automated data quality rule generation (LangGraph) from Snowflake-only to all supported databases, then built an evaluation framework to measure prompt changes against real-world data. Managed 5 releases of the open-source library (18k+ GitHub stars) with bug fixes, new metrics, and reliability improvements including zero-downtime schema migrations and message queue thread-safety fixes.