AI EngineeringPython

PerfUp

by raullenchai

PerfUp is an AI Engineering skill for Claude Code, published by raullenchai in Rapid-MLX.

3.5K stars398 forkson raullenchai/Rapid-MLXAdded 2026/08/16+1% in starsRepository updated 2026/08/16
apple-siliconclaude-codecursordeepseekfastapihacktoberfestinferencellmlocal-llmm1m2m3macosmlxollama-alternativeopenai-apipythonqwentool-calling
Install in seconds
Install PerfUp
Copy PerfUp into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.
terminal
npx degit https://github.com/raullenchai/Rapid-MLX/tree/main/.claude/skills/perfup ~/.claude/skills/perfup

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/raullenchai/Rapid-MLX.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file
.claude/skills/perfup/SKILL.md in raullenchai/Rapid-MLX
Installs to
~/.claude/skills/perfup
Collection
The only skill cataloged from this repository
Category
AI Engineering2451 skills

What PerfUp does

PerfUp autonomously researches, benchmarks, and implements performance optimizations for vllm-mlx, then ships production-ready pull requests when improvements are confirmed.

PerfUp is cataloged under AI Engineering on DirSkills. PerfUp comes from a repository tagged apple-silicon, claude-code, cursor, deepseek and fastapi.

Documentation

README

/perfup — Autonomous Performance Optimization

Inspired by karpathy/autoresearch: you are an autonomous performance researcher for vllm-mlx. You propose optimizations, benchmark them, keep what works, discard what doesn't, and ship a production PR.

Key Files

  • Results log: reports/perfup-results.tsv — append-only experiment log (commit, metric, status, description)
  • Optimization queue: memory/knowledge/perf_optimization_queue.md — ranked list of candidates
  • Memory index: memory/MEMORY.md — what's been done, what's known
  • Benchmark script: scripts/benchmark_engines.py
  • Model for benchmarking: Check memory for current model path. If unavailable, ask user.

This is the opening of the README. Read the full README on GitHub.

Commands PerfUp provides

Slash commands named in this skill’s SKILL.md, listed in the order they first appear.

  • /perfup
  • /review-loop

Frequently asked about PerfUp

  • How does PerfUp compare to other AI Engineering skills?

    PerfUp ranks #593 by stars among the 2451 AI Engineering skills in this catalog. The most-starred ones next to it are Architecture Decision Records, AI-First Engineering and Agentic OS. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of PerfUp against them. Open each page to compare what they document and how they install.