Django Perf Review

What it does

This skill analyzes Django application performance metrics sourced from Sentry, focusing on identifying and validating potential bottlenecks within the codebase. It aims to pinpoint specific areas for optimization by tracing data flow and verifying issues before reporting them. The tool prioritizes finding provable problems and avoids generating false positives.

When to use it

• When investigating slow request times or timeouts in a Django application.
• To proactively identify performance bottlenecks before they impact users at scale.
• During code reviews to ensure efficient database queries and data access patterns.
• After deploying new features or making significant changes to the codebase.
• When Sentry reports indicate potential performance issues requiring deeper investigation.

Key capabilities

• N+1 Query Detection: Identifies inefficient query patterns where a loop iterates through results, triggering an additional database query for each item.
• Unbounded Queryset Analysis: Detects potentially memory-exhausting querysets that could lead to "Out of Memory" (OOM) errors.
• Missing Index Identification: Highlights opportunities to improve query performance by adding missing indexes on large tables.
• Write Loop Detection: Identifies loops causing lock contention and slowing down request processing.
• Prioritization based on Impact: Focuses on critical and high-impact issues (e.g., N+1 queries, unbounded querysets) that significantly affect application performance at scale.

Example prompts

• "Analyze the Django code for potential N+1 query patterns."
• "Review recent Sentry reports for performance bottlenecks in the user list view."
• "Identify any missing indexes contributing to slow database queries."

Tips & gotchas

• The skill relies on data from Sentry; ensure proper instrumentation and reporting are configured.
• Validation is crucial – the tool flags potential issues, but manual verification of the reported problems is required.
• Focus on critical and high-impact findings first to maximize efficiency.