performance-optimizer

To resolve API latency, establish a baseline, use distributed tracing to isolate bottlenecks like N+1 database queries or missing indexes, apply one targeted fix, and verify the result against the baseline.

Diagnosing memory leaks requires profiling real workloads to isolate CPU hotspots and memory issues, establishing a reproducible baseline, and validating one controlled change at a time to eliminate the leak without regressions.

Optimizing Core Web Vitals involves measuring frontend loading and interaction metrics, profiling real user workloads, applying targeted improvements, and enforcing performance budgets via CI checks to prevent regressions.

Yes, you can use percentile-based measurements to define performance budgets and SLOs, interpreting distributed tracing evidence to ensure production readiness and prevent latency regressions in CI pipelines.

Database performance degradation often stems from unoptimized queries, missing indexes, or connection pooling issues, requiring profiling evidence and targeted optimization to resolve bottlenecks under concurrency.

One-change-at-a-time validation limits concurrent optimization testing, requiring reproducible baselines and profiling evidence to accurately measure each isolated fix's impact on the overall performance.