skills.rest
fanguyun
fanguyunCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

ml-pipeline

Design ML pipelines with experiment tracking, orchestration, feature stores, and registries.

Updated May 31, 2026
Search Tags:#mlops#ml pipeline#experiment tracking#model registry#feature store#training orchestration
One-click install
npx skills add https://github.com/fanguyun/SkillManager --skill ml-pipeline-fanguyun
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ml-pipeline
Source: https://github.com/fanguyun/SkillManager/tree/main/ml-pipeline
Command: npx skills add https://github.com/fanguyun/SkillManager --skill ml-pipeline-fanguyun

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires mlflow, kubeflow, airflow, prefect, feast, mlflow, weightsandbiases, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill helps design, implement, and automate production-grade machine learning pipelines, reducing manual labor and ensuring best practices are followed.

Core Features & Use Cases

  • Experiment Tracking: Configure and manage experiment tracking with MLflow, Weights & Biases, or custom solutions.
  • Pipeline Orchestration: Set up and manage Kubeflow, Airflow, or Prefect DAGs for training orchestration.
  • Feature Store: Build and maintain feature store schemas with Feast or custom solutions.
  • Model Registry: Deploy model registries and automate retraining and validation workflows.
  • Use Case: Imagine you need to build a ML pipeline for a classification model. This Skill will guide you through setting up experiment tracking, creating a training pipeline, deploying the model, and automating the retraining process.

Quick Start

Run the ml-pipeline skill with the command: ml-pipeline init

Frequently Asked Questions about ml-pipeline

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a production-grade ML pipeline with MLflow and Airflow?▼

Build a production-grade ML pipeline by configuring MLflow for experiment tracking and orchestrating training DAGs with Airflow to ensure reproducibility, versioning, and automated validation workflows.

What is needed to set up a feature store for machine learning pipelines?▼

Setting up a feature store requires Feast or a custom solution to define and maintain feature schemas. This centralizes feature engineering, ensures consistency between training and serving, and supports pipeline automation.

Can I use Prefect or Kubeflow for ML training orchestration?▼

Yes, you can use Prefect or Kubeflow for ML training orchestration. They manage Directed Acyclic Graphs (DAGs) to automate complex training workflows, schedule jobs, and handle dependencies effectively.

How do I automate model retraining and validation workflows?▼

Automate model retraining and validation workflows by integrating a model registry with orchestration tools like Kubeflow or Airflow. This triggers scheduled retraining jobs based on metrics and validates performance automatically.

Does this MLOps approach support Weights & Biases for experiment tracking?▼

Yes, this MLOps approach supports Weights & Biases for experiment tracking. You can configure and manage experiments using W&B alongside MLflow to monitor metrics, compare model versions, and facilitate collaboration.

What is the best way to manage ML model registries and versioning?▼

The best way to manage ML model registries and versioning is using MLflow. It tracks model iterations, stores artifacts, and integrates with orchestration pipelines to automate deployment and retraining workflows.

Related Skills

Explore More Data & Analytics Skills→
shalevaminshalevamin

ml-pipeline

Designs ML pipeline infrastructure automating model lifecycle and integrating MLflow or Feast.

Community
Advanced
89jobrien89jobrien
4

machine-learning

Automate the machine learning lifecycle from experimentation to production deployment.

Community
Advanced
Design-System-ETDesign-System-ET

ml-pipeline

Design Kubeflow or Airflow DAGs for ML training workflows with MLflow tracking.

Official
Advanced
karrtik159karrtik159

mlops-engineer

Build and manage Kubernetes-based ML pipelines with experiment tracking and model versioning.

Community
Advanced