performance-review

To check code changes for scalability risks, run a targeted performance scan on workspace diffs or pull requests to surface suspicious hotspots, query patterns, and resource usage before production deployment.

An N+1 query risk occurs when looped database access triggers excessive individual queries. Performance signal scanning identifies these cost drivers in your code changes and recommends batching or pagination to mitigate latency.

Create a benchmark plan by converting performance scan findings into concrete scale assumptions, then selecting appropriate verification methods like load testing, profiling, or explain plans to observe production metrics and validate latency behavior.

Yes, you can assess caching strategy and pagination risks for a specific branch refactor by applying worst-case reasoning to the diff, evaluating cache invalidation needs and pagination requirements to prevent production incidents.

The best way to review database query patterns is using deterministic signal scanning on feature diffs to identify cost mechanisms, followed by generating actionable mitigation guidance like adding indexes or streaming to optimize resource usage.

Automated performance review focuses on cost mechanisms like query patterns and resource usage, so it may not suit changes lacking measurable scalability impact, such as pure UI adjustments or documentation updates with no latency risk.