226 skills · sorted by relevance
Optimize AgentDB performance with quantization (4-32x memory reduction), HNSW indexing (150x faster search), caching, and batch operations. Use when optimizing memory usage, improving search speed, or scaling to millions of vectors.
The English summary is being prepared. Showing the original description.
Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.
The English summary is being prepared. Showing the original description.
Achieve aggressive v3 performance targets: 2.49x-7.47x Flash Attention speedup, 150x-12,500x search improvements, 50-75% memory reduction. Comprehensive benchmarking and optimization suite.
The English summary is being prepared. Showing the original description.
Optimize AgentDB performance with quantization (4-32x memory reduction), HNSW indexing (150x faster search), caching, and batch operations. Use when optimizing memory usage, improving search speed, or scaling to millions of vectors.
The English summary is being prepared. Showing the original description.
Performance optimization guidelines for OneKey React/React Native applications. Use when optimizing app performance, fixing UI freezes/lag, reducing re-renders, handling concurrent operations, or analyzing performance bottlenecks. Triggers on performance, optimization, optimize, slow, lag, freeze, hang, jank, stutter, memory, leak, concurrent, batching, batch, memoization, memo, bridge, windowSize, contentVisibility, FlashList, re-render, fps, tti, bundle.
The English summary is being prepared. Showing the original description.
Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.
The English summary is being prepared. Showing the original description.
Write efficient C code with proper memory management, pointer arithmetic, and system calls. Handles embedded systems, kernel modules, and performance-critical code. Use PROACTIVELY for C optimization, memory issues, or system programming.
The English summary is being prepared. Showing the original description.
Analyze application and system performance to find bottlenecks, the points that slow processing. Use resource usage and performance data to plan improvements to resource allocation, load balancing, caching, and algorithms, then compare results after changes.
Analyze slow development workflows and agent coordination using execution times, resource use, dependencies, and communication patterns. Identify bottlenecks, prioritize causes, and build an improvement plan covering parallel work, caching, or resource allocation.
Build Claude applications using guidance for the official Anthropic SDK. Choose an integration approach, implement streaming and tool use, and follow workflows for model migration, prompt audits, evaluation, and cost optimization.
Build Python workflows for filtering, grouping, joining, and reshaping table data with Polars. Get guidance on migrating pandas code and using lazy evaluation, which optimizes a processing plan before execution, for large datasets.
Review CLAUDE.md files, which provide project instructions to Claude Code. Receive a scored quality report covering commands, structure, and outdated guidance, then approve focused edits that keep the instructions concise and useful.
Compress LLM weights using HQQ quantization, which stores weights with fewer bits without calibration datasets. Follow workflows for precision settings, model saving, inference integration, and LoRA fine-tuning of quantized models.
Inspect Google Cloud Filestore usage and free space, then propose capacity increases or decreases based on configured thresholds and storage tier limits. Apply resizing after confirmation to address low space or reduce excess capacity.
Adapt large language models to specific tasks using LoRA and QLoRA, methods that train a small set of added parameters. Get guidance on preparing examples, choosing training settings, evaluating outputs, and saving or merging trained adapters.
Improve PostgreSQL queries and database design by examining execution plans and choosing indexes for actual access patterns. Get guidance on large tables, JSONB data, connection pooling, and maintenance while accounting for write overhead and memory use.
Triage incorrect MNN outputs and regressions by matching symptoms to focused debugging guides. Use backend comparisons, minimal reproductions, and controlled experiments to investigate memory issues, quantization, concurrency, numeric limits, caches, and inconsistent results across runs.
Apply HQQ quantization, which stores language model weights using fewer bits without calibration examples. Covers precision settings, inference backends, saving compressed models, deployment with HuggingFace Transformers or vLLM, and further training with LoRA.
Convert language models to GGUF and reduce their size through quantization, which stores model weights at lower precision. Guides cover compression choices, local inference on CPUs and Apple Silicon, and integration with tools such as Ollama.
Clean, combine, summarize, and reshape tabular data with pandas DataFrames, Python's table-like data structures. Produces transformation code and validation checks, including time series processing and memory-conscious approaches for large datasets.
Adapt large language models to specific tasks through fine-tuning, or additional training. Prepares and validates datasets, configures LoRA or QLoRA to train a small subset of parameters, and supplies training settings, evaluation scripts, and deployment optimization guidance.
Diagnose slow systems, crashes, and resource problems on servers, PCs, virtual machines, and containers. Quick checks identify affected components, while deeper analysis investigates causes and provides optimization advice or repair plans.
Measure and improve Three.js or WebGL game performance through repeatable gameplay checks. Identify CPU or GPU bottlenecks, reduce unnecessary rendering and processing, and verify that controls, collisions, visuals, and memory remain stable.
Estimate whether a model execution fits the available hardware and identify likely performance limits. Separate memory capacity, computation, data movement, and sequential waiting to produce a table of assumptions, ideal minimum times, unknowns, and measurement priorities.