Browse skills

Build multi-metric SEO/GEO stakeholder reports - executive summary, organic traffic, rankings, GEO/AI citations, authority, and content quality - each metric source-tagged and compared to a prior period.

Fast structured generation and serving for LLMs with RadixAttention prefix caching. Use for JSON/regex outputs, constrained decoding, agentic workflows with tool calls, or when you need 5× faster inference than vLLM w…

Builds conversion funnels from raw event data with drop-off attribution, segment comparison, and significance testing, and delivers a filled funnel table with a diagnosed bottleneck and next action. Use when someone asks "where are users dropping off", "why did checkout conversion fall this week", "build a signup-to-activation funnel", or "is mobile converting worse than desktop". Do NOT use for rewriting page copy or layout to lift a known weak step - use landing-page-cro instead; for decomposing active-user growth into new, retained, resurrected, and churned - use growth-accounting instead; for choosing which events to instrument and which product metrics to track - use product-analytics instead; for designing and reading a controlled experiment - use ab-test-analyzer instead.

Design and automate surveys, benchmarks, and data collection to create unique competitive advantages through original research.

Designs presentation tables that read correctly at a glance - numeric alignment, consistent precision, deliberate sort order, unit-labeled headers, and well-placed totals - and delivers a before/after redesign of the reader's table. Use when someone asks "make this table more readable", "how many decimals should I show", "should totals go at the top or bottom", "this report table is a mess", or is preparing tabular data for an exec deck or published report. Do NOT use for designing database tables, keys, and normalization - use database-schema instead; for narrative structure around the numbers - use data-story instead; for assembling the full multi-section document - use formatted-report-writer instead.

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 …

Produces a rigorous, honest evaluation report for an ML model - real baselines, business-matched metrics with confidence intervals, slice-level error analysis, calibration checks, and a go/no-go recommendation. Use when someone asks "is this model good enough to ship", "write up the eval for this model", "did we actually beat the old model", or before any model promotion, stakeholder demo, or retraining decision. Do NOT use for evaluating LLM or prompt outputs - use llm-evaluation instead; for analyzing randomized product experiments use ab-test-analyzer; for documenting a shipped model for consumers use model-card-writer.

Multi-agent orchestration framework for autonomous AI collaboration. Use when building teams of specialized agents working together on complex tasks, when you need role-based agent collaboration with memory, or for pr…

Apply prioritized PostgreSQL query, connection, schema, concurrency, RLS, access, and monitoring practices.

Framework for state-of-the-art sentence, text, and image embeddings. Provides 5000+ pre-trained models for semantic similarity, clustering, and retrieval. Supports multilingual, domain-specific, and multimodal models.…

Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for redu…

Evaluates LLMs across 100+ benchmarks from 18+ harnesses (MMLU, HumanEval, GSM8K, safety, VLM) with multi-backend execution. Use when needing scalable evaluation on local Docker, Slurm HPC, or cloud platforms. NVIDIA'…