Analyze ML training and inference code to identify bottlenecks and propose optimizations.
npx skills add https://github.com/tim-krausz/mlstack --skill review-perf-tim-krausz
Please help me install this Agent Skill. Skill: review-perf Source: https://github.com/tim-krausz/mlstack/tree/main/review-perf Command: npx skills add https://github.com/tim-krausz/mlstack --skill review-perf-tim-krausz
Analyze and optimize ML training and inference code by locating bottlenecks in GPU utilization, data throughput, memory footprint, and overall throughput, then provide concrete optimization guidance with measurable impact.
Provide the ML project or script bundle and I will deliver a performance review with concrete changes and a step-by-step optimization plan.
High-intent search queries and answers about installing and using this skill.
ML pipeline profiling analyzes training scripts, data loaders, and inference code to locate GPU utilization, data throughput, and memory bottlenecks, then provides prioritized optimization plans with measurable impact estimates and implementation guidance.
Improving GPU utilization requires recommending optimized mixed precision modes, data formats, loading strategies, and batching configurations that increase pipeline speed or reduce cost without sacrificing model correctness across common ML frameworks.
Profiling a training script involves applying structured profiling steps to a provided ML project bundle, which generates a reproducible report containing before and after metrics alongside a step-by-step optimization checklist.
Yes, the performance review process works with inference pipelines and profiling notebooks by analyzing the code to identify bottlenecks and propose concrete optimizations applicable across common ML frameworks.
To run a baseline performance review, you need to provide the ML project or script bundle, which is analyzed to deliver a performance review with concrete changes and a step-by-step optimization plan.
ChuaHanChongRun evaluation commands and profile GPU usage to establish ML model baseline metrics.
BFLabsAIIdentify and prioritize Postgres performance improvements for Supabase deployments.
yangwhaleInstall, configure, and troubleshoot DeepEP on NVIDIA GPU systems.
albibenniConfigure ESLint, Prettier, and TypeScript settings based on detected frameworks and package managers.
