skills.rest
yanochka11
yanochka11COMMUNITY
Explore all skills by this author →
GitHub

perf-profiler

Profile Python applications and ML training to diagnose performance bottlenecks.

Updated May 5, 2026
Search Tags:#performance#python#optimization#memory#profiling#gpu
One-click install
npx skills add https://github.com/yanochka11/harness_bro --skill perf-profiler-yanochka11
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: perf-profiler
Source: https://github.com/yanochka11/harness_bro/tree/main/.claude/skills/curated/perf-profiler
Command: npx skills add https://github.com/yanochka11/harness_bro --skill perf-profiler-yanochka11

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps diagnose slow software, excessive memory usage, inefficient GPU utilization, and performance bottlenecks by applying targeted profiling instead of guesswork.

Core Features & Use Cases

  • Performance Profiling: Selects appropriate tools such as cProfile, py-spy, scalene, and torch.profiler based on the observed performance issue.
  • Bottleneck Analysis: Measures runtime or memory behavior, identifies the top resource consumers, and guides focused optimization changes.
  • Use Case: When a machine learning training job runs slowly or a Python service leaks memory, use this Skill to profile the workload, locate the bottleneck, and validate improvements with before-and-after measurements.

Quick Start

Use the perf-profiler skill to analyze why my Python training script is slow and recommend measured optimization steps.

Frequently Asked Questions about perf-profiler

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I profile a slow Python training script to find the bottleneck?▼

To profile slow Python code, you select appropriate tools like cProfile, py-spy, scalene, or torch.profiler to measure runtime behavior, identify top resource consumers, and guide focused optimization changes.

What is the best way to diagnose excessive memory usage in a Python application?▼

Diagnosing excessive memory usage requires targeted profiling to measure memory behavior, locate the bottleneck causing the leak, and validate optimization improvements using quantitative before-and-after measurements.

Can I use profiling to fix inefficient GPU utilization during machine learning workflows?▼

Yes, profiling fixes inefficient GPU utilization in machine learning workflows by measuring resource usage, applying tools like torch.profiler to locate bottlenecks, and validating optimization improvements with quantitative results.

Does performance profiling require specific tools for different Python optimization tasks?▼

Yes, performance profiling requires selecting appropriate tools like cProfile, py-spy, scalene, or torch.profiler based on the observed issue, ensuring you accurately measure runtime, memory, or GPU bottlenecks.

How do I validate performance optimization improvements after fixing a bottleneck?▼

Validate performance optimization improvements by measuring runtime or memory behavior before and after changes, ensuring you quantitatively confirm the impact of fixes applied to the identified bottleneck.

Related Skills

Explore More Software Engineering Skills→
NVIDIA-AI-IOTNVIDIA-AI-IOT
110

jetson-memory-audit

Measure Jetson DRAM and NvMap usage on live NVIDIA Jetson devices.

Official
Advanced
ihmorolihmorol
1

optimize

Identify performance bottlenecks and implement optimizations with benchmark verification.

Community
Advanced
DevelopersGlobalDevelopersGlobal
65

observability

Implement structured logging, distributed tracing, and alerting with OpenTelemetry.

Official
Advanced
Michae2xlMichae2xl

python-performance-optimization

Profile Python code with cProfile and memory_profiler to identify bottlenecks.

Community
Advanced