Data Science & ML Claude 스킬

Pixi Package Manager

Manages fast, reproducible scientific Python environments by unifying conda and PyPI ecosystems within a single workflow.

PDF Text Extraction & Conversion

Extracts and converts PDF content into clean, LLM-ready markdown or text using AI-powered and high-fidelity local tools.

Excel Analysis & Data Automation

Streamlines the analysis, transformation, and visualization of Excel spreadsheets using advanced Python libraries within Claude Code.

Logit Lens Interpretability

Decodes intermediate transformer layer activations into human-readable vocabulary predictions to visualize model processing.

RAG Architecture Systems Design

Design and optimize high-performance Retrieval-Augmented Generation (RAG) pipelines for LLM applications.

Large Corpus Fan-Out Analyzer

Orchestrates a multi-stage subagent pipeline to analyze large datasets or codebases that exceed standard context windows.

AI Bias Assessment & Mitigation

Evaluates AI systems for fairness and implements mitigation strategies using demographic parity, equalized odds, and proxy detection.

Prompt Engineering Specialist

Designs, evaluates, and optimizes high-performance LLM prompts using systematic engineering patterns and rigorous testing frameworks.

Data Science Strategy & Brainstorming

Refines vague data analysis requests into clear, actionable objectives through Socratic questioning and structured specification.

DS Fix: Data Science Troubleshooting

Provides a structured diagnostic framework for fixing broken data science workflows, notebook errors, and incorrect analysis results.

Disease Mechanism Entry Creator

Automates the creation of scientifically accurate disease mechanism entries in the Dismech knowledge base using a research-first approach.

VRAM & GPU Memory Manager

Manages GPU VRAM allocation through OOM retry logic, idle auto-unloading, and cross-service signaling protocols.

Recursive Arena Orchestrator

Refines complex AI outputs through iterative multi-model battles and consensus-driven self-critique.

Data Commons Client

Accesses and analyzes global public statistical data from the Data Commons knowledge graph for research and development.

Scientific Critical Thinking

Evaluates research rigor by assessing methodology, experimental design, and statistical validity using frameworks like GRADE and Cochrane.

Letta AI Agent Framework

Simplifies building and managing stateful AI agents with long-term memory using the Letta framework.

Biopython Molecular Biology Toolkit

Performs advanced computational molecular biology tasks including sequence analysis, database queries, and structural bioinformatics.

Bioinformatics Services (BioServices)

Provides programmatic access to over 40 bioinformatics web services and databases for streamlined biological data retrieval and analysis.

Astropy Astronomy & Physics

Performs advanced astronomical data analysis, coordinate transformations, and cosmological calculations using the Astropy Python library.

Multi-Model Battle & Refinement

Orchestrates iterative refinement cycles across multiple LLMs to produce high-quality, peer-reviewed outputs through competitive judging.

PyDESeq2 RNA-Seq Analysis

Performs comprehensive differential gene expression analysis on bulk RNA-seq data using the Python implementation of DESeq2.

scvi-tools

Analyzes single-cell omics data using deep generative models and probabilistic frameworks for genomics research.

GRPO RL Fine-Tuning Expert

Implements Group Relative Policy Optimization (GRPO) for training language models in reasoning, logic, and structured output tasks.

RAG Implementation Expert

Implements robust Retrieval-Augmented Generation systems using vector databases and semantic search to ground AI responses in external knowledge.

AgentDB Reinforcement Learning

Trains autonomous AI agents using nine reinforcement learning algorithms and WASM-accelerated neural inference for rapid behavioral optimization.

Flow Nexus Neural Networks

Trains and deploys distributed neural networks within E2B sandboxes for scalable AI model development and orchestration.

Reactome Biological Pathway Analysis

Automates querying the Reactome pathway database for gene enrichment, molecular interactions, and systems biology research.

Scientific Python Packaging

Streamlines the creation of distributable scientific Python packages using modern pyproject.toml standards and community-best practices.

System Prompt Architect

Generates production-grade, structured system prompts by analyzing complex user requirements and defining optimal AI agent architectures.

Human Metabolome Database (HMDB) Access

Accesses and analyzes data from the Human Metabolome Database for metabolite identification, biomarker discovery, and clinical research.