Data Science & ML Claudeスキル

Embedding Optimization

Optimizes vector embedding pipelines for RAG systems through efficient model selection, strategic chunking, and cost-effective caching.

AI Data Engineering

Builds production-ready data infrastructure for AI systems, including RAG pipelines, feature stores, and embedding workflows.

Prompt Engineering

Engineers reliable, high-quality LLM outputs using advanced techniques like few-shot learning, chain-of-thought reasoning, and structured data patterns.

MLOps Implementation Strategy

Provides strategic guidance and technical patterns for operationalizing machine learning models from experimentation to production deployment.

Model Serving & LLM Deployment

Deploys and optimizes LLM and machine learning models for production-grade inference and API integration.

SaaS AI Feature Expert

Integrates cost-effective AI capabilities into SaaS applications through optimized prompt engineering, RAG patterns, and strategic API selection.

Single-Cell Deep Learning (scvi-tools)

Streamlines single-cell genomics workflows using scvi-tools for probabilistic data integration and deep learning analysis.

Instrument Data to Allotrope Converter

Converts laboratory instrument data into standardized Allotrope Simple Model (ASM) formats for LIMS integration and data analysis.

Nextflow Bioinformatics Pipelines

Automates the deployment and execution of nf-core bioinformatics pipelines for genomics, transcriptomics, and epigenomics data analysis.

Scientific Research Strategy

Guides scientists through a systematic framework for research problem selection, project ideation, and strategic decision-making.

Single-Cell RNA Quality Control

Automates quality control workflows for single-cell RNA-seq data using scverse best practices and MAD-based outlier detection.

Clinical Trial Protocol Designer

Generates comprehensive clinical trial protocols for medical devices and drugs using a structured, research-driven automated workflow.

Allotrope Instrument Data Converter

Converts diverse laboratory instrument files into standardized Allotrope Simple Model (ASM) format for LIMS and data analysis.

Nextflow Bioinformatics Development

Automates the deployment and management of nf-core bioinformatics pipelines for large-scale genomic data analysis.

scvi-tools Single-Cell Analysis

Accelerates single-cell genomics research using probabilistic deep learning models for data integration, multi-modal analysis, and reference mapping.

Single-Cell RNA-seq Quality Control

Performs automated quality control and filtering on single-cell RNA-seq data using scverse best practices.

Neuro-Symbolic Reasoning

Integrates large language models with symbolic logic solvers to perform complex reasoning, theorem proving, and constraint satisfaction tasks.

Large Corpus Fan-Out Analysis

Orchestrates a parallel multi-agent pipeline to analyze large text or code corpora that exceed a single agent's context window.

FinLab Quant Trading

Automates the development, backtesting, and execution of Taiwan stock market quantitative trading strategies using the FinLab package.

Metadata Spell Checker

Detects and corrects spelling errors in ETL metadata and snapshot files using codespell.

Denario Scientific Research Assistant

Automates end-to-end scientific research workflows from hypothesis generation to publication-ready LaTeX manuscripts.

Qdrant Java Integration

Integrates Qdrant vector database with Java and Spring Boot applications using LangChain4j for advanced semantic search and RAG capabilities.

LangChain4j Spring Boot Integration

Integrates LangChain4j with Spring Boot to build production-ready AI applications using declarative services and automated configuration.

LangChain4J Vector Store Configurator

Configures and optimizes high-performance vector database integrations for Java-based RAG and semantic search applications.

LangChain4j RAG Patterns

Implements Retrieval-Augmented Generation (RAG) systems using LangChain4j to build AI applications with external knowledge access.

Prompt Engineering & Optimization

Optimizes LLM performance through advanced prompt patterns, few-shot learning, and structured chain-of-thought reasoning frameworks.

Amazon Bedrock Java SDK Patterns

Integrates Amazon Bedrock foundation models into Java and Spring Boot applications using the AWS SDK for Java 2.x.

RAG Implementation Kit

Provides standardized patterns and best practices for building Retrieval-Augmented Generation (RAG) systems with vector databases and semantic search.

RAG Chunking Strategy

Implements sophisticated document segmenting techniques to optimize Retrieval-Augmented Generation (RAG) performance and vector search accuracy.

LangChain4j AI Services Patterns

Simplifies the integration of large language models into Java applications using declarative interfaces and annotations.