Performance Review

Identify performance bottlenecks in the selected code and suggest concrete optimizations.

What to look for

Algorithmic complexity

• O(n²) or worse where a better algorithm exists
• Linear scans on data that should be indexed or hashed
• Repeated work that could be computed once

Memory

• Memory leaks (event listeners, timers, closures holding references)
• Excessive allocations in hot paths
• Large objects held longer than needed

I/O and concurrency

• Sequential awaits that could run in parallel
• Blocking operations on the main thread / event loop
• Missing connection pooling or reuse

Database

• N+1 queries — fetching related data in a loop
• Missing indexes on filtered/sorted columns
• SELECT * when only a few columns are needed
• Missing query result caching for expensive, stable queries

Frontend (if applicable)

• Unnecessary re-renders (missing memoization, unstable references)
• Large bundle chunks loaded eagerly
• Unoptimized images or missing lazy loading

Output format

For each issue:

• Where: file and line
• What: the bottleneck
• Impact: rough estimate (if measurable)
• Fix: corrected code or approach