Open-Source

Between the commit and the query, between the review and the resolution. An open-source contribution to git-lrc — the AI code-review tool that hit #3 Product of the Day on Product Hunt. Designed and built lrc query, a SQL-based engine in Go that turns a repository's entire code-review history into queryable, structured data — extraction to in-memory SQLite to table or JSON output in a single pipeline. Benchmarked at linear scale, 150ms per 100k commits, with range-bounding logic added to keep large monorepos from choking the engine. Traced execution flows across multiple Go modules, translated issue reports into production-ready fixes, and matched the project's test coverage standards across multiple review cycles with the maintainer. AI-assisted analysis layered over manual debugging to surface architectural loopholes, edge-case failures, and workflow bugs — each finding translated into a concrete, actionable issue or a merged improvement.