Browse our curated library of 2,600+ SKILL.md files for AI agents. Download ready-to-use skills for Claude Code, Codex, OpenAI and more.
Designs and runs robust LLM evaluation frameworks and benchmarks to reveal strengths, weaknesses, and actionable improvement paths.
Guides targeted LLM fine-tuning, selecting methods (LoRA, adapters) and data strategies to adapt models with minimal forgetting.
Optimizes LLM inference and serving to meet SLAs at lower cost via profiling, batching, KV cache, quantization, and autoscaling.
Provides expert guidance on LLM pretraining, data curation, tokenizer design, and distributed training infrastructure.
Explain structure–property–processing relationships in engineering materials and guide selection, processing routes, and performance optimization for real-world applications.
This skill guides ML cost optimization across training and inference, including GPU right-sizing, caching, and FinOps for measurable savings.
Guides deploying ML models and building MLOps pipelines for production, covering serving, registry, monitoring, and automated retraining.
This skill provides senior MLOps guidance to configure and optimize experiment tracking (W&B, MLflow, Neptune, ClearML) for reproducible runs.
Guides rigorous ML evaluation, metric selection, baselining, and monitoring to prevent production failures.
Guides selecting the right machine learning model for a problem, baselining with simple options and matching data, constraints, and timelines.
Senior MLOps expert for production ML monitoring, including drift detection, ground-truth collection, performance and prediction monitoring, dashboards, alerts, and SLA management.
Guides end-to-end ML pipeline design and MLOps, emphasizing reproducible, observable, and maintainable production systems.
Explain MOBI/AZW Kindle ebook formats, their structure, DRM variants, and conversion workflows for creating and distributing Kindle-compatible ebooks.
Architect scalable mobile analytics by defining event taxonomies, selecting analytics stacks, and guiding data-driven product decisions.
A senior MLOps expert guiding model registry design, versioning, lifecycle management, and governance across platforms such as MLflow, Weights & Biases, and SageMaker.
Guidance on designing and operating production ML model serving infra (TorchServe, Triton, TF Serving, BentoML) for online, batch, or streaming inference, APIs, and autoscaling.
Build observability with metrics, logs, and traces to understand system behavior and diagnose issues.
Expert guidance on multimodal deep learning, vision-language models, and cross-modal representations for tasks like CLIP, VQA, and image captioning.
Expert guidance on neural architecture search, efficient design, and model compression for automating NAS, one-shot searches, and hardware-aware optimizations.
Guides designing neural network architectures by selecting layers, connections, and hyperparameters to fit data and tasks; use when building or adapting models.
Design and evaluate end-to-end NLP pipelines from ingestion to output for tasks like classification, NER, and QA.
Architect and evaluate no-code applications, selecting platforms, modeling data, and designing scalable internal tools without code.
Understand PDF structure, standards (ISO 32000-2), and practical tasks like viewing, parsing, and converting PDFs across platforms.
Profile applications to identify bottlenecks, measure performance, and optimize hot paths to meet targets and SLAs.
Guides apps through Apple App Store and Google Play compliance, preventing violations, handling rejections, and crafting effective appeals.