Provides an end-to-end methodology for integrating ML into existing non-ML codebases, covering problem scoping, data readiness, architecture decoupling, and baseline model integration.
How this skill is triggered — by the user, by Claude, or both
Slash command
/everything-claude-code:ml-adoption-playbookThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
此技能提供一种自适应方法,用于把机器学习模型实现到现有软件工程项目中。它通过结构化 ML 应如何研究、解耦、训练和集成,连接传统 SWE 与 MLOps。
此技能提供一种自适应方法,用于把机器学习模型实现到现有软件工程项目中。它通过结构化 ML 应如何研究、解耦、训练和集成,连接传统 SWE 与 MLOps。
写模型代码前,先明确“为什么”和“怎么做”。
没有干净、可访问的数据,ML 没用。
不要把模型推理紧耦合到核心业务逻辑。
fastapi-patterns 或 django-patterns),或放进专用 service class。让代码可复现、可迭代。
pytorch-patterns 或类似最佳实践:固定 random seeds、让代码 device-agnostic、明确记录 tensor/array shapes。基线模型集成后,把重点转向持续运维。
mle-workflow: 引导用户设置 experiment tracking、model registries 和 drift detection。使用此 playbook 协助用户时,智能体应:
npx claudepluginhub aaione/everything-claude-code-zhProvides a structured methodology for adding machine learning to existing non-ML codebases, covering problem framing, data readiness, architectural decoupling, and baseline model integration.
Turns model work into production ML systems with data contracts, repeatable training, quality gates, deployable artifacts, and monitoring. Useful for ranking, search, recommendations, classifiers, forecasting, embeddings, LLMs, anomaly detection, and batch analytics.
Optimizes AI/ML/LLM usage in production systems via usage audits, model selection, prompt engineering, cost modeling, A/B experiments, and data pipelines.