301 skills · sorted by relevance
Calculate financial ratios from company statements to assess profitability, ability to cover obligations, efficiency, and valuation. Results include explanations, comparisons across periods, available industry benchmarks, and a formatted Excel report.
Review an AI workflow across five areas: prompts, use of context, tools, system structure, and safety. Receive scores, specific findings, and prioritized actions to address quality and reliability problems.
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.
Integrate Flash Attention into transformer models to reduce processing time and GPU memory use for long sequences. Covers PyTorch integration, advanced attention options, and benchmarks that compare speed, memory use, and output accuracy.
Review recent code changes for duplication, maintainability, and wasted work, then apply straightforward improvements. The workflow combines three parallel reviews, prioritizes small, low-risk fixes, and checks the edited code with relevant tests or project checks.
Route customer success, sales, revenue operations, and contract requests to one of four specialist skills. Use it when a business request is unclear, so the agent selects the relevant workflow before starting the work.
Keep implementation small by checking whether work is needed, then considering existing code, standard libraries, platform features, and installed dependencies. Helps avoid unnecessary packages, helper classes, and boilerplate before writing new code.
Reviews Google Cloud spending and resource usage to identify waste, recommend suitable resource sizes, compare discount options, and establish ongoing cost management using the cost optimization guidance in the Google Cloud Well-Architected Framework.
Assess Google Cloud workloads for environmental sustainability using the Well-Architected Framework. Get recommendations on lower-carbon regions, efficient computing and storage, AI energy use, and ongoing emissions measurement with Carbon Footprint.
Review BigQuery capacity, table layouts, storage retention, and SQL to improve cost efficiency. Compare billing options, plan partitioned tables, and evaluate search indexes, stored aggregation results, and BI Engine for suitable workloads.
Prepare for engineering interviews with structured practice in algorithms, system design, and questions about past experience. Develop explanations of your reasoning, analyze solution efficiency, discuss design tradeoffs, and organize concrete stories about your work.
Design evaluation systems that measure agent results across accuracy, completeness, citations, and tool efficiency. Build repeatable tests, compare changes against a baseline, and define quality checks that catch regressions before deployment and monitor production performance.
Plan how an AI agent uses limited input space through reusable prompt prefixes, shorter tool outputs, summaries, and task splitting. Measure token use, response quality, and latency to choose optimizations that provide a net benefit.
Learn how instructions, documents, tool results, and conversation history form an AI model’s context. Understand attention limits, why information in the middle can be missed, and how relevance and placement inform decisions about what to include.
Design persistent agent memory that retains knowledge across sessions and retrieves relevant facts when needed. Compare storage approaches, track people and changing information, and plan memory consolidation and evaluation based on the kinds of questions the agent must answer.
Optimize transformer attention, the calculation that relates tokens to one another, using Flash Attention. The skill guides PyTorch integration, GPU memory reduction, performance measurement, and output comparisons for training or inference with long sequences.
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.
Configure large language model training across multiple NVIDIA GPUs with Megatron-Core. The skill helps choose how to split models and workloads, set training parameters, monitor performance, and address memory shortages or low GPU utilization.
Design and improve prompts for language models, then test their accuracy and consistency. Create reusable templates, JSON output schemas that define response structure, evaluation criteria, and test cases while considering token use and model differences.
Investigate observed slowdowns or increased CPU, memory allocation, and IO costs in a process you control. Use controlled timing and runtime-specific profiling tools to test a cause, then check performance and behavior after a fix.
Review Apache Pinot changes affecting frequently executed code or making performance claims in a pull request. Relevant areas include operators, segment readers, aggregation functions, and locking inside loops that process individual rows.
Rewrite long prompts into shorter, clearer instructions for language models. Preserve essential requirements while removing repetition, defining scope, and making output formats and success criteria explicit to reduce ambiguity and unnecessary tokens.
Comprehensive Google Ads account health analysis detecting wasted spend, search term leaks, negative keyword gaps, bid strategy issues, and Quality Score problems. Use when analyzing campaign data, pasting Google Ads exports, reviewing account performance, or requesting a full diagnostic of advertising spend efficiency. Platform: Google.
The English summary is being prepared. Showing the original description.
Multi-channel budget optimization using MER, marginal ROAS, and diminishing returns analysis. Use when pasting multi-channel spend and results data, requesting reallocation recommendations, analyzing budget shift priorities, or optimizing marketing efficiency across Google, Meta, TikTok, and other channels. Platform: Google and Meta.
The English summary is being prepared. Showing the original description.