mercari/ml-system-design-pattern resource

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

2926 stars · 340 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A documentation collection of system design patterns for building, serving, and operating machine learning systems in production, published by Mercari as a GitHub Pages site. It covers serving patterns, QA/evaluation patterns, and antipatterns, with sample implementations in a companion repository.

Use cases

• design a model serving architecture for production
• learn MLOps system design patterns
• choose between synchronous and asynchronous prediction serving
• set up A/B testing for machine learning models
• avoid common ML system antipatterns
• deploy ML models on Kubernetes or cloud

When to choose

• you are architecting ML systems for production deployment
• you want reference designs for model serving, monitoring, and QA
• you need language-agnostic ML infrastructure guidance

When to avoid

• you need runnable production code rather than design documentation
• you are looking for patterns to improve model accuracy
• you need actively updated content - the repository has not seen releases since 2021

Facets

learning-resource · maturity maintenance

documentation machine-learning deployment monitoring testing machine-learning tutorials cloud-computing cloud python ml-system-design mlops design-patterns model-serving production-ml architecture-patterns devops architecture kubernetes

Member repositories

For agents

markdown · JSON · MCP: product_card(name="mercari/ml-system-design-pattern")