Does this skill help with model versioning?

Yes, it integrates best practices for model registries like MLflow and Weights & Biases, along with data versioning strategies using tools like DVC.

How does it handle pipeline failures?

The skill includes patterns for modularity, idempotency, and sophisticated error handling, including retry logic and automated alerting for failed stages.

What orchestration tools does this skill support?

The skill provides guidance and templates for industry-standard tools including Apache Airflow, Dagster, Kubeflow Pipelines, and Prefect.

Can I use this for real-time inference pipelines?

While it excels at batch training workflows, it also covers real-time feature pipelines and model serving patterns suitable for low-latency production environments.

MLOps Pipeline Workflow

Name: MLOps Pipeline Workflow
Author: HermeticOrmus

byHermeticOrmus

0•

Data Science & ML

Orchestrates end-to-end machine learning lifecycles from data preparation and training to production deployment and monitoring.

This skill provides a comprehensive framework for building and orchestrating robust MLOps pipelines. It guides users through creating Directed Acyclic Graphs (DAGs) for workflow management, implementing automated data validation, managing experiment tracking, and establishing reliable deployment strategies like canary or blue-green releases. It is an essential tool for data scientists and ML engineers looking to transform manual experimental code into scalable, reproducible, and production-ready automated systems with built-in observability and failure handling.

Key Features

01Production-grade serving and deployment strategies

02Experiment tracking and model registry integration

03End-to-end DAG orchestration patterns for ML workflows

040 GitHub stars

05Automated data validation and feature engineering pipelines

06Automated performance monitoring and drift detection

Use Cases

01Designing reproducible training pipelines for new ML models

02Implementing CI/CD practices specifically for machine learning systems

03Automating the transition from raw data ingestion to production serving

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add hermeticormus/alqvimia-contador ml-pipeline-workflow

For use in Claude.ai and ChatGPT

Download Skill