performance-optimization

To resolve backend performance bottlenecks, establish a response time baseline, profile where time is spent across queries or network calls, apply a targeted minimal fix to that specific area, and verify improvements with before and after measurements.

Diagnose frontend performance regressions by capturing traces and waterfalls to attribute main thread delays to specific LCP, INP, or CLS metrics, then applying minimal code changes limited to the identified rendering bottleneck for verification.

The best way to prove performance optimization results is to measure a baseline before applying changes, implement minimal targeted fixes, and produce a verified before and after comparison demonstrating measurable improvement on the slow path.

Set up a regression guard by defining performance thresholds during your optimization workflow, then adding monitoring to continuously track metrics against the established baseline to prevent future slowdowns on the critical path.

Yes, this approach applies to both frontend rendering bottlenecks and backend latency issues, requiring baseline creation, bottleneck attribution, and minimal targeted changes limited to the specific slow area before verifying results.

You should not apply broad refactoring during performance optimization because the methodology requires minimal code changes limited strictly to the identified bottleneck area to ensure improvements are attributable and measurable.