skills.rest
romankovsv
romankovsvCOMMUNITY
Explore all skills by this author →
GitHub

mlops

Design end-to-end MLOps architectures for scalable ML pipelines.

Updated Mar 29, 2026
Search Tags:#mlops#dvc#feature-store#experiment-tracking#sagemaker#feast#model-drift
One-click install
npx skills add https://github.com/romankovsv/claude-code-python-devops-mlops --skill mlops-romankovsv
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: mlops
Source: https://github.com/romankovsv/claude-code-python-devops-mlops/tree/main/skills/mlops
Command: npx skills add https://github.com/romankovsv/claude-code-python-devops-mlops --skill mlops-romankovsv

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Data science and MLOps teams need a cohesive framework to build, reproduce, and scale end-to-end ML pipelines spanning data versioning, experiment tracking, feature stores, and model deployment.

Core Features & Use Cases

  • End-to-end ML lifecycle guidance: from data versioning (DVC), experiment tracking (MLflow / W&B), to feature stores (Feast vs AWS Feature Store) and model deployment (SageMaker, Seldon Core).
  • Drift detection and monitoring: design pipelines with automated monitoring, drift checks, and retraining triggers to maintain model accuracy over time.
  • Use Case: architect a production ML platform that ensures reproducibility, governance, and seamless deployment across cloud-native serving runtimes.

Quick Start

Demonstrate an end-to-end MLOps workflow from data versioning through model deployment using SageMaker and Seldon Core.

Frequently Asked Questions about mlops

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

FAQPage Schema
How do I build an end-to-end MLOps pipeline for production ML models?▼

An end-to-end MLOps pipeline integrates data versioning with DVC, experiment tracking via MLflow, feature stores like Feast, and model deployment using Seldon Core to ensure reproducibility and scalability across production environments.

What is the best way to detect model drift and trigger automated retraining?▼

Detecting model drift requires designing pipelines with automated monitoring and drift checks that trigger automated retraining, maintaining model accuracy over time without manual intervention.

Does this MLOps architecture support deploying models with SageMaker and Seldon Core?▼

Yes, this MLOps architecture supports deploying models using SageMaker and Seldon Core, enabling seamless deployment across cloud-native serving runtimes while ensuring governance and reproducibility.

How do I use DVC and MLflow together for data versioning and experiment tracking?▼

Using DVC and MLflow together combines data versioning with experiment tracking, allowing teams to version datasets while logging model parameters and metrics to ensure reproducible ML experiments.

When do I need a feature store like Feast in my ML pipeline?▼

You need a feature store like Feast when building production ML platforms that require consistent feature serving, governance, and seamless integration across training and inference workflows.

Related Skills

Explore More Data & Analytics Skills→
Yangel-hideYangel-hide

lambda-labs-gpu-cloud

Provision on-demand GPU instances and clusters via the Lambda Cloud API.

Community
Intermediate
arsityarsity

serving-llms-vllm

Deploy vLLM servers with OpenAI-compatible endpoints and tensor parallelism.

Community
Advanced
weebcoder101weebcoder101
3

devops

Manage Docker container builds and GitHub Actions CI/CD pipeline configurations.

Community
Advanced
handsomelong922handsomelong922

serving-llms-vllm

Deploy scalable LLM serving with vLLM for OpenAI-compatible endpoints.

Community
Advanced