Data Science & ML Claudeスキル

Python Anti-Slop

Enforces production-grade Python standards by eliminating generic AI coding patterns through PEP 8 compliance, rigorous type hinting, and pandas best practices.

Prompt Engineering Patterns

Optimizes LLM interactions through advanced prompting techniques like few-shot learning, chain-of-thought, and systematic template design.

Pixi Package Manager

Manages fast, reproducible scientific Python environments by unifying the conda and PyPI ecosystems using the Rust-based Pixi tool.

Data Anomaly Detection

Identifies outliers and unusual patterns in datasets using advanced machine learning algorithms to uncover fraud, defects, or security threats.

TimeGPT-2 Migration Assistant

Automates the transition of time-series forecasting pipelines from TimeGPT-1 to the enhanced TimeGPT-2 architecture.

MAIC Biostatistics Methodology

Provides expert guidance and best practices for conducting Matching-Adjusted Indirect Comparisons (MAIC) in biostatistical research.

AI & ML Prompt Optimizer

Optimizes Large Language Model (LLM) prompts to minimize token consumption, reduce operational costs, and enhance response quality.

Code Analyzer

Analyzes GitHub repositories to extract computational methodologies and automatically draft scientific Methods sections.

PyMC 5 Fundamentals

Facilitates the creation, review, and optimization of Bayesian models using PyMC 5 and ArviZ diagnostics.

TimeGPT Pipeline Builder

Generates production-ready Python forecasting pipelines using Nixtla's TimeGPT API to automate complex time-series analysis.

Nixtla Forecast Validator

Validates time series forecast quality and detects performance degradation by comparing current metrics against historical benchmarks.

Nixtla Experiment Architect

Scaffolds production-ready time-series forecasting experiments using Nixtla's suite of machine learning and statistical libraries.

STC Methodology for Biostatistics

Guides the implementation and review of Simulated Treatment Comparisons (STC) using NICE DSU TSD 18 compliant methodologies.

RAG Systems Implementation

Builds robust Retrieval-Augmented Generation (RAG) systems for LLM applications using vector databases and semantic search.

Feature Engineering Toolkit

Automates the creation, selection, and transformation of data features to optimize machine learning model performance and accuracy.

Arxiv Paper Reader

Accesses and analyzes Arxiv research papers directly within Claude to facilitate academic research and technical Q&A.

Recommendation Engine Builder

Builds sophisticated recommendation systems using collaborative filtering, content-based, and hybrid modeling techniques.

Machine Learning Model Trainer

Automates the end-to-end process of training, evaluating, and persisting machine learning models from raw datasets.

BUGS/JAGS Bayesian Modeling

Simplifies the creation and review of Bayesian models using BUGS and JAGS declarative syntax and precision parameterization.

RAG Implementation Toolkit

Builds robust Retrieval-Augmented Generation (RAG) systems using vector databases and semantic search to ground AI responses in proprietary knowledge.

LangChain Application Architect

Architects sophisticated LLM applications using the LangChain framework to implement agents, memory management, and complex workflow chains.

Teradata SVM Classification

Automates the end-to-end Support Vector Machine (SVM) workflow for linear and non-linear classification tasks within Teradata Vantage.

Time Series Forecaster

Automates time series analysis and forecasting to predict future trends and seasonal patterns using advanced machine learning models.

ML Integration Helpers

Deploys machine learning models into full-stack applications using production-ready FastAPI endpoints, Next.js UI components, and Supabase schemas.

OpenRouter Model Routing Patterns

Implements intelligent model routing strategies to optimize for cost, performance, and reliability when building AI applications with OpenRouter.

Prompt Engineering Patterns

Optimizes LLM performance and reliability through advanced prompting techniques like few-shot learning and chain-of-thought reasoning.

Cloud GPU Configurator

Configures cloud GPU environments and provides selection guidance for Modal, Lambda Labs, and RunPod platforms.

MLOps Pipeline Orchestrator

Build, automate, and manage end-to-end MLOps pipelines from data ingestion through production deployment.

Nixtla Uncertainty Quantifier

Quantifies forecast uncertainty by generating prediction intervals and confidence bands for time series models using conformal prediction.

RAG Chunking Strategies

Implements and benchmarks advanced document chunking strategies to optimize retrieval performance and context preservation in RAG pipelines.