Browse Skills

A meta-skill that makes AI coding assistants think strategically before acting. Implements a 5-factor risk heuristic (Reversibility, Blast Radius, Coupling, Stakes, Uncertainty) to determine whether a code change needs shallow execution, medium analysis, or deep pre-mortem reasoning. Based on cognitive science, game theory, and software engineering economics research.

Master XGBoost for tabular data problems including hyperparameter optimization, feature engineering, SHAP-based interpretability, and production deployment with drift monitoring.

Comprehensive prompt engineering guidance covering techniques (chain-of-thought, few-shot, system prompts), evaluation, guardrails, structured output, and building reliable AI interactions.

Comprehensive guide to Claude Code CLI covering skills, MCP servers, hooks, slash commands, project setup, CLAUDE.md configuration, and workflows for maximizing AI-assisted development.

Systematically removes LLM fingerprints from text using evidence-based linguistic research. Replaces machine patterns (vocabulary, structure, tone) with human writing characteristics while preserving meaning and quality.

Designs production-grade Retrieval-Augmented Generation pipelines including chunking strategies, embedding model selection, vector database architecture, hybrid search, reranking, and hallucination reduction techniques.

Design reliable structured data extraction and generation systems powered by LLMs, with JSON Schema design, provider-specific configuration, validation pipelines, and production error handling.

Design intelligent model routing systems that optimize cost, latency, and quality across multiple LLM providers. Covers cascading strategies, confidence-based escalation, prompt caching economics, and fallback architectures.

Systematic optimization of machine learning model hyperparameters using Bayesian optimization, multi-fidelity methods, and distributed search strategies to find optimal configurations efficiently.

Builds structured competitive intelligence analyses including competitor profiling, feature parity matrices, win/loss patterns, pricing intelligence, and sales battlecards using proven CI frameworks and market research methodologies.

Guides the design of reinforcement learning systems including policy selection (PPO, SAC, DQN), environment design, reward shaping, exploration strategies, multi-agent configurations, and sim-to-real transfer for robotics and game AI.

Design safe, effective tool-using AI agents with proper tool schemas, agent loop architecture, safety gates, and evaluation strategies for production LLM applications.

Ensure offline/online feature consistency in ML systems by designing validation pipelines, detecting training-serving skew, and implementing point-in-time correctness guarantees across feature store architectures.

Expert architect for designing multi-step prompt chains and orchestration patterns — from task decomposition and chain-of-thought sequencing to intermediate verification gates, context window management, error recovery strategies, and production-grade prompt pipeline design for complex LLM workflows.

Comprehensive guide to integrating LLMs into applications covering API usage, RAG, agents, evaluation, cost optimization, streaming, and production deployment patterns.

Design, structure, and optimize production-grade system prompts for large language models using proven architectural patterns, research-backed techniques, and adversarial testing methodologies.

Design and implement structured deep learning training workflows using PyTorch Lightning, covering LightningModule architecture, distributed training strategies, experiment tracking, checkpoint management, and production model deployment.

Design and optimize production-grade NLP pipelines covering text preprocessing, tokenization, NER, sentiment analysis, and transformer-based architectures using spaCy and HuggingFace.

Designs recommendation systems covering collaborative filtering, content-based filtering, hybrid approaches, cold start solutions, embedding models, real-time serving, and A/B testing strategies for personalized user experiences.

Designs production-grade drift monitoring pipelines for ML models, covering data drift, concept drift, and prediction drift detection with statistical methods, alerting thresholds, and automated response procedures.

Expert guidance for building production-grade LLM applications with LangChain. Covers chain composition with LCEL, RAG pipelines, agent systems, LangGraph workflows, prompt engineering, and production deployment with LangSmith observability.

Helps you set up and optimize MLflow for experiment tracking, run comparison, model registry management, and production MLOps workflows across any ML framework.

Designs comprehensive evaluation frameworks for LLM and AI systems, combining automated metrics (BLEU, ROUGE, BERTScore), LLM-as-judge patterns (G-Eval), RAG evaluation (RAGAS), standard benchmarks (MMLU, HumanEval), safety evaluations, and A/B testing methodologies into production-ready evaluation pipelines.

Guides decisions on when and how to fine-tune large language models, covering LoRA/QLoRA, dataset curation, RLHF/DPO alignment, evaluation strategies, and cost-performance tradeoffs for production deployments.