Representative MoE training playbooks by hardware platform and model family. Summarizes rounded throughput bands, parallelism patterns, and common tuning stacks.
Systematic workflow for MoE training optimization in Megatron Bridge, based on the Megatron-Core MoE paper.
Analyze ncu (NVIDIA Nsight Compute) profiling output: SOL% bottleneck classification, roofline analysis, occupancy diagnosis, memory hierarchy analysis, warp stall analysis,…
Nsight Systems (nsys) CLI for system-level timeline profiling. Use when the user wants to run nsys profile, analyze .nsys-rep reports, use nsys stats/analyze/recipe commands,…
Performance optimization coordination playbook. Contains specialist routing table, TileIR two-step pipeline, kernel generation specialist selection, prioritization criteria, and…
Operational guide for choosing and combining parallelism strategies in Megatron Bridge, including sizing rules, hardware topology mapping, and combined parallelism config — from…
Hotspot detection and performance root-cause analysis — flamegraph interpretation, allocation tracking, latency profiling, regression measurement.
Use for CPU profiling on macOS or Linux — samply, Instruments, perf, py-spy, counters, flame graphs.
Catch the algorithmic and I/O performance classics before shipping — O(n²) on unbounded input, N+1 queries, sync calls in hot/async paths, unbounded memory.
Validate and use packed sequences and long-context training in Megatron-Bridge, distinguishing offline packed SFT for LLMs from in-batch packing for VLMs, and applying th — from…
Use when generating performance hypotheses backed by git history and code evidence. — from general/general-misc
Operational guide for enabling TP, DP, and PP communication overlap in Megatron-Bridge, including config knobs, code anchors, pitfalls, and verification. — from NVIDIA/skills
Session-after-session product perfection loop for Vibeman. The strongest available model (Fable) directs — it walks the repo's context map context-by-context, proposes 5…
Perfect Gym Solutions S.A. integration. Manage data, records, and automate workflows. Use when the user wants to interact with Perfect Gym Solutions S.A. data.
Autonomous but risk-bounded options trading agent spec for the "Perfect Storm" strategy (paper trading only).
AI 애니메이션 스프라이트 생성. 텍스트 설명 한 줄로 캐릭터 + 동작 애니메이션(걷기·달리기·공격·마법 등 100여 종) + 8방향 스프라이트 세트를 만들고, 게임 엔진용 번들(스프라이트시트 · manifest.json · Aseprite JSON · 상태별 GIF/APNG · 개별 프레임 PNG)로 내보냅니다.
Analyze Android, Linux, and Chromium Perfetto traces with local trace_processor_shell evidence. Use for startup, scrolling or jank, input latency, ANR, CPU scheduling, memory or…
API de Perfis de Clientes da Tray. Utilize quando o desenvolvedor precisar gerenciar perfis (grupos/segmentos) de clientes da loja, incluindo criação, consulta, atualização,…
Design and evaluate perforation programs for oil and gas wells including conventional single-zone and multi-cluster horizontal completions.
Comprehensive performance analysis, bottleneck detection, and optimization recommendations for Claude Flow swarms — from Claude Flow Team
Comprehensive performance analysis, bottleneck detection, and optimization recommendations for Claude Flow swarms — from bg-szy/TOP-SKILLS
Automatically analyze performance issues when user mentions slow pages, performance problems, or optimization needs.
Use when writing code that runs per item, per frame, or per event, or building a cache or lookup over a collection that can grow large.
Decompose portfolio returns into explainable components to identify where value was added or lost. Use when the user asks about Brinson attribution, allocation vs selection…
Comprehensive performance analysis and Core Web Vitals optimization for Drupal and WordPress projects. Spawns performance-specialist for full analysis.
Capture performance baselines before migration for regression comparison and SLA verification
Establish performance baselines and catch regressions (load time, Web Vitals, size).
Performance budget tanımı + CI enforcement. Web Vitals (LCP/INP/CLS), JS/CSS bundle size, API p95/p99 latency, DB query, memory/CPU, cost budget.
Use when declaring, measuring, or enforcing performance thresholds as a quality contract rather than as an aspirational target.
Use when the user wants to pull live metrics from all connected analytics platforms and get an instant performance snapshot with KPIs, trends, and comparisons to targets.
Correlate content attributes with performance metrics across GA4, GSC, and SE Ranking. Identify what drives performance and build optimization hypotheses.
Evidence gathering for performance review cycles. Gathers goal completion evidence, peer feedback, development progress, scope changes, and values alignment, organised along the…
Runs the performance review on the design draft, producing grounded findings against the 50k req/s target.
Diagnoses pipeline performance issues -- slow jobs, expensive queries, latency trends -- using Monte Carlo's cross-platform observability.
Operate as a performance engineer who sets budgets, gates regressions in CI, and runs profiling as a service other teams rely on.
Expert in system optimization, profiling, and scalability. Specializes in eBPF, Flamegraphs, and kernel-level tuning.
Use when optimizing speed, memory, latency, startup time, throughput, token cost, build time, or runtime resource use.
Performance specialist perspective for the weekly review. Focuses on bundle size, LCP / CLS / INP, avoidable re-work, image and font optimization.
Auto-invoke when reviewing loops, data fetching, rendering, database queries, or resource-intensive operations.
Verify performance implications were considered and no obvious anti-patterns exist. Issues result in WARNINGS.
Diagnose and fix iOS performance — launch time, main-thread hangs and scroll hitches, memory growth and leaks, and energy, using Instruments, os_signpost, and MetricKit field…
Use when the main problem is latency, throughput, memory, resource usage, render slowness, or another performance bottleneck that needs investigation and prioritization.
Performance Lighthouse Runner - Auto-activating skill for Frontend Development. Triggers on: performance lighthouse runner, performance lighthouse runner Part of the Frontend…
Record actual post-publish performance data — likes, retweets, PV, saves, comments, shares, and revenue per content piece.
Use when when you have access to a set of gallery or benchmark scripts executed across multiple plotting backends and need to quantify which backend delivers the fastest median…
Log task completion data to .claude/metrics/. Use at the end of every task to record tokens, cost, agents used, rework cycles, and hallucination events.
Real-time performance monitoring and metrics collection for HR-IMS | การตรวจสอบประสิทธิภาพแบบเรียลไทม์และการเก็บรวบรวมเมตริกสำหรับ HR-IMS
Expert in observing, benchmarking, and optimizing AI agents. Specializes in token usage tracking, latency analysis, and quality evaluation metrics.
Optimizes application performance. Use when performance requirements exist, when you suspect performance regressions, or when Core Web Vitals or load times need improveme — from…
React Native(Expo) 앱의 성능을 최적화하는 스킬. 번들 크기, 렌더링, 리스트 가상화, 이미지, 네트워크, 앱 시작 시간을 최적화한다. 앱 성능 최적화, 속도 개선, 메모리 최적화 시 반드시 이 스킬을 사용.
Identifies and fixes performance bottlenecks in code, databases, and APIs. Measures before and after to prove improvements.
Identifies and fixes performance bottlenecks. Use when optimizing code performance, fixing slow operations, or when asked to improve speed.
Use when investigating slowness with baselines, profiler evidence, and ranked bottlenecks. Triggers:
Identifies performance bottlenecks including N+1 queries, inefficient loops, memory leaks, and slow algorithms.
Performance profiling principles. Measurement, analysis, and optimization techniques.
Gates สำหรับตรวจจับ performance regression ผ่าน benchmarks, bundle size, และ load time monitoring
Create Chinese work reports from weekly reports, git commits, repository metrics, notes, and reusable templates while preserving template formatting.
Berliner Start-up-HR: Performance Reviews: faire Kriterien, Bias-Check, Dokumentation, Gehaltsbezug, Entwicklungsplan und Low-Performer-Risiko.
Drafts and critiques performance reviews using Glenn Kramon's seven rules from Stanford GSB's Winning Writing course.
Write self-assessments, prepare for review conversations, calibrate against rubrics, and "manage up" the formal review process.