QueryLens is a PostgreSQL observability system that streams query telemetry, classifies regressions, and surfaces them in a backend plus frontend stack.
- The project spans backend collection, regression analysis, and dashboard presentation.
- It includes a C++ collector and a React UI.
- The repo already contains regression evaluation documentation.
- Query telemetry -> collector -> backend API -> regression detection/reporting -> frontend dashboard.
- Use the repo-specific
maketargets documented inREADME.md - Frontend development via
npm run devinfrontend/
- Backend tests via
pytest - Frontend build via
npm run build
- Review
docs/REGRESSION_EVALUATION.md - Review
docs/BENCHMARKS.md - Review
docs/RECOMMENDATIONS.md - Future benchmark artifacts should live under
benchmarks/
- No canonical benchmark summary was extracted in this pass.
- Benchmark generation is now packaged into a runner, but live results still depend on a local database and backend.
- The seed-demo story is stronger now, but it still relies on fixture-backed data.
- Add a compact benchmark summary doc with parsed artifacts.
- Built a PostgreSQL observability stack that detects query regressions from streamed telemetry.
- Combined backend collection, regression analysis, and dashboard visualization in one system.
- Documented seeded regression evaluation workflows for reproducibility.
python3 /Users/sushildalavi/Desktop/Github/QueryLens/scripts/run_benchmark.py --pending --events 10000 --workers 4- pass - 2026-06-17 - Wrote pending JSON and Markdown artifacts underbenchmarks/.python3 /Users/sushildalavi/Desktop/Github/QueryLens/scripts/run_benchmark.py --pending --events 100 --workers 2 --output-dir /tmp/querylens-bench --artifact-name querylens_test- pass - 2026-06-17 - Verified the pending artifact path with a custom output location.- Direct helper checks against
scripts/run_benchmark.py- pass - 2026-06-17 - Verified pending and live artifact shapes through direct assertions.