Can I use this for LLM application development?

Absolutely. It includes 2025 best practices for LLM queries, structured output extraction, retry logic, and advanced prompt engineering techniques.

Does this skill support both deep learning and classical ML?

Yes, it includes comprehensive implementation patterns for Scikit-Learn for traditional machine learning and PyTorch for deep learning architectures.

How does it handle data pipeline failures?

The skill provides specialized data pipeline patterns with checkpointing and recovery mechanisms to allow for resumption of expensive transformations after a failure.

Does it include model evaluation and tracking?

Yes, it integrates MLOps best practices using MLflow for experiment tracking, parameter logging, and performance metric evaluation.

Data & AI Engineering

Name: Data & AI Engineering
Author: pluginagentmarketplace

bypluginagentmarketplace

•

データサイエンスとML

Builds and deploys robust machine learning models, data pipelines, and intelligent LLM applications using Python and industry-standard frameworks.

概要

This skill empowers Claude to act as a senior Data Scientist and AI Engineer, providing expert guidance across the entire machine learning lifecycle. It covers everything from fundamental data manipulation with NumPy and Pandas to advanced deep learning with PyTorch, MLOps with MLflow, and the integration of large language models. Whether you are building predictive models, designing robust data pipelines with checkpointing, or implementing sophisticated prompt engineering, this skill ensures your Python code follows industry best practices, includes proper validation, and incorporates modern error-handling patterns for production-grade AI systems.

主な機能

Integrated MLOps practices including experiment tracking and model logging with MLflow
Comprehensive testing templates and troubleshooting guides for common ML and CUDA issues
Scalable data engineering pipelines with built-in checkpointing and recovery
1 GitHub stars
End-to-end Machine Learning lifecycle management using Scikit-Learn and PyTorch
Production-grade LLM application development with retry logic and structured outputs

ユースケース

Building RAG-capable applications and LLM-powered agents with robust error handling
Designing and training predictive models for classification, regression, or ranking tasks
Developing automated data processing pipelines that handle large datasets efficiently

概要

主な機能

Integrated MLOps practices including experiment tracking and model logging with MLflow
Comprehensive testing templates and troubleshooting guides for common ML and CUDA issues
Scalable data engineering pipelines with built-in checkpointing and recovery
1 GitHub stars
End-to-end Machine Learning lifecycle management using Scikit-Learn and PyTorch
Production-grade LLM application development with retry logic and structured outputs

ユースケース

Building RAG-capable applications and LLM-powered agents with robust error handling
Designing and training predictive models for classification, regression, or ranking tasks
Developing automated data processing pipelines that handle large datasets efficiently