2.7K skills · sorted by relevance
Choose a model training method and configuration from paired response preferences or individual positive and negative ratings. Covers DPO, ORPO, KTO, and SimPO, including pair construction, repeated training rounds, and troubleshooting learning settings.
Investigate slow SQL using execution plans, which show how a database processes a query. Review indexes, filters, joins, and maintenance practices through PostgreSQL examples to identify bottlenecks and plan changes to queries or database structure.
Improve slow Apache Spark jobs by reviewing data partitions, caching, memory settings, and shuffles, which move data between workers. Use the guidance to investigate bottlenecks and tune pipelines for larger datasets.
Review DAX formulas, the expressions used for calculations in Power BI, and propose clearer, more maintainable alternatives. Identify repeated calculations, costly filtering, and error-handling issues, then explain changes and recommend ways to test them.
Improve LLM prompts using Arize production traces, evaluation results, and human feedback. Extract current instructions, identify recurring failures, revise the prompt, and compare experiments to measure changes in output quality and catch regressions.
Analyze AWS infrastructure definitions, deployed resources, usage metrics, and actual costs to identify savings opportunities. Rank recommendations by savings, risk, and effort, then create individual GitHub issues and a coordinating issue after approval.
Measure denoising latency in SGLang diffusion models and identify slow operations with profiler traces. Compare performance and output quality on the same GPUs, check existing optimizations, and gather evidence before starting low-level kernel work.
Develop Azure Functions with guidance for .NET, Python, and Node.js. Covers asynchronous processing, Durable Functions for coordinating tasks, startup delay optimization, and production concerns such as timeouts, HTTP client reuse, and monitoring configuration.
Improve PostgreSQL performance through a staged workflow covering slow query analysis, indexes that speed up data access, query rewrites, configuration tuning, routine maintenance, and monitoring for production databases.
Plan and improve popups and banners for email signups, purchases, and announcements. The skill helps define display timing, audiences, copy, and frequency, with mobile design guidance and comparison tests to evaluate changes.
Create blog posts and social media content with a consistent brand voice, meaning a recognizable writing style. Includes tools for analyzing existing text and suggesting search optimization improvements, plus templates for adapting content and planning publication calendars.
Reduce the memory needed to run large language models with GPTQ quantization, which stores model weights at lower precision. Covers loading compressed models, quantizing your own models, choosing execution backends, and connecting to memory-efficient fine-tuning.
Improve mobile app listings on Apple App Store and Google Play Store through keyword research, clearer descriptions, visual asset guidance, and review analysis. Create competitor reports, A/B test plans, and launch checklists to guide store page improvements.
Train language models with SimPO using pairs of preferred and rejected answers. Configure preference training without a separate reference model, adapt settings for base or instruction models, and address unstable training, memory limits, or lost capabilities.
Get guidance on adapting LLMs with Unsloth, including implementation, API usage, and debugging. Covers LoRA and QLoRA, methods for adjusting models with fewer trainable parameters, with a focus on training speed and memory use.
Compress large language models to 4-bit weights to reduce GPU memory use and accelerate inference. Covers loading compressed models, configuring quantization, and using Transformers or vLLM, with guidance on hardware requirements and backend choices.
Configure machine learning training and batch jobs across multiple clouds with SkyPilot. The skill guides GPU selection, lower-cost cloud and region choices, and automatic recovery when discounted spot instances are interrupted.
Reduce GPU memory use by loading language models in 8-bit or 4-bit formats with bitsandbytes and HuggingFace Transformers. Includes configuration examples, QLoRA fine-tuning of compressed models, and optimizers that reduce memory needed during training.
Write, review, and refactor React and Next.js code using performance guidelines ranked by impact. Covers sequential data requests, downloaded code size, server work, and unnecessary rendering to guide concrete implementation improvements.
Write, review, or refactor React code using prioritized guidance for data fetching, bundle size, rendering, component structure, testing, and type safety. Start with high-impact issues such as sequential requests and unnecessary library loading.
Review code for performance bottlenecks across database queries, memory use, frontend rendering, network requests, and algorithms. Use a structured checklist to identify optimization opportunities and classify their severity according to expected load and scale.
Plan popups that encourage email signups, purchases, or inquiries. Define when they appear, who sees them, how often they repeat, and what they say, with mobile adjustments, overlap rules, and ideas for comparing alternatives.
Audit and improve content intended for citation in AI answers. Review evidence, author information, and structure, generate revised content and structured metadata, and keep a local record of observed citations across AI services.
Produce blog posts, articles, and guides from research through a finished draft. Build an outline from audience needs and search intent, gather sources, then refine readability, brand voice, search optimization, links, and metadata for publication.