skills.rest
irahardianto
irahardiantoCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

ml-engineering

Design reproducible ML pipelines from feature engineering through deployment and monitoring.

150|48|Updated Jan 24, 2026
Search Tags:#deployment#ml pipeline#feature engineering#MLOps#experiment tracking#model validation#model monitoring
One-click install
npx skills add https://github.com/irahardianto/awesome-agv --skill ml-engineering-irahardianto
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ml-engineering
Source: https://github.com/irahardianto/awesome-agv/tree/main/.agents/skills/ml-engineering
Command: npx skills add https://github.com/irahardianto/awesome-agv --skill ml-engineering-irahardianto

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the risk of fragile, non-reproducible machine learning systems that fail in production due to poor pipeline design, weak validation, and missing operational monitoring.

Core Features & Use Cases

  • ML pipeline design: Define end-to-end stages from data collection through deployment and monitoring, with guardrails for reproducibility and experiment tracking.
  • Feature engineering discipline: Prevent training-serving skew using reusable feature transformations and maintain feature documentation for freshness and lineage.
  • Validation and MLOps monitoring: Catch leakage and performance regressions early while tracking drift in data, features, and model quality over time.

Quick Start

Use the ml-engineering skill to design an end-to-end training-to-serving plan with a validation checklist, a monitoring strategy for drift, and a tool stack for experiment tracking, feature storage, and model registry.

Frequently Asked Questions about ml-engineering

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

FAQPage Schema
How do I design an ML pipeline that prevents training-serving skew?▼

To prevent training-serving skew, design ML pipelines with reusable feature transformations and maintain feature documentation for freshness and lineage. This ensures feature engineering discipline across both training and serving environments.

What is the best way to monitor machine learning models for data drift in production?▼

Monitoring machine learning models for data drift requires tracking changes in data, features, and model quality over time. Implement automated retraining triggers to catch performance regressions early and maintain operational reliability.

How do I build reproducible ML systems for production deployment?▼

Building reproducible ML systems requires end-to-end pipeline design from data collection through deployment with versioned artifacts and experiment tracking. This prevents fragile systems that fail due to poor pipeline design and weak validation.

Does this MLOps approach support real-time inference and batch serving patterns?▼

This MLOps approach supports production serving patterns including batch, real-time, streaming, and edge inference. It applies evaluation criteria across these deployment patterns while tracking model performance drift.

How do I validate machine learning models to avoid data leakage?▼

Validating machine learning models to avoid data leakage requires evaluation against performance and fairness criteria. Implement validation checklists during the training-to-serving transition to catch leakage early.

Why do I need experiment tracking in my ML pipeline?▼

Experiment tracking in your ML pipeline ensures reproducibility by maintaining versioned artifacts throughout the training-to-serving workflow. It prevents fragile systems that fail in production due to missing operational monitoring.

Related Skills

Explore More Data & Analytics Skills→
ManfronEnricoManfronEnrico

Feature Engineering

Engineer features from raw tabular data with encoding, scaling, and generation.

Community
Advanced
aiunlocked1412aiunlocked1412
14

ml-engineer

Design and deploy production ML systems with MLOps pipelines for serving and monitoring.

Community
Advanced
nksethnkseth

ml-pipeline

Automate ML pipeline construction with Kubeflow, Airflow, MLflow, Feast, and DVC.

Community
Advanced
ekremmkasapekremmkasap

ml-pipeline-workflow

Orchestrate end-to-end ML pipelines from data preparation to deployment.

Community
Advanced