Skip to content
BeClaude

ml-engineering

New
18.2kGitHubGeneralby stas00

Machine Learning Engineering Open Book

Summary

This skill provides an open-book reference for machine learning engineering, covering best practices, design patterns, and common pitfalls in building, deploying, and maintaining ML systems.

  • It helps developers quickly recall key concepts, code snippets, and architectural decisions without searching the web.

Install & Usage

1
Create the skills directory
mkdir -p .claude/skills
2
Download the skill file

Add the configuration to .claude/skills/ml-engineering.md

3
Invoke in Claude Code
/ml-engineering

Use Cases

Designing a scalable data pipeline for training a deep learning model on large datasets.
Selecting the appropriate model serving infrastructure (e.g., TensorFlow Serving, TorchServe) for a production application.
Debugging common issues in distributed training such as gradient synchronization or data loading bottlenecks.
Implementing MLOps practices like experiment tracking, model versioning, and automated retraining.
Choosing between feature stores and online/offline feature computation for real-time inference.
Optimizing inference latency and throughput using techniques like quantization, pruning, or batching.

Usage Examples

1

/ml-engineering How do I set up a distributed training job on Kubernetes using PyTorch DDP?

2

/ml-engineering What are the trade-offs between using a feature store vs. computing features on the fly for a recommendation system?

3

/ml-engineering Show me a code snippet for implementing gradient accumulation in TensorFlow to handle large batch sizes.

View source on GitHub

Security Audits

LicenseUnknownSourceWarnRepositoryPass

Frequently Asked Questions

What is ml-engineering?

This skill provides an open-book reference for machine learning engineering, covering best practices, design patterns, and common pitfalls in building, deploying, and maintaining ML systems. It helps developers quickly recall key concepts, code snippets, and architectural decisions without searching the web.

How to install ml-engineering?

To install ml-engineering: create the skills directory (mkdir -p .claude/skills), then add the config to .claude/skills/ml-engineering.md. Finally, /ml-engineering in Claude Code.

What is ml-engineering best for?

ml-engineering is a other categorized under General. Created by stas00.

What can I use ml-engineering for?

ml-engineering is useful for: Designing a scalable data pipeline for training a deep learning model on large datasets.; Selecting the appropriate model serving infrastructure (e.g., TensorFlow Serving, TorchServe) for a production application.; Debugging common issues in distributed training such as gradient synchronization or data loading bottlenecks.; Implementing MLOps practices like experiment tracking, model versioning, and automated retraining.; Choosing between feature stores and online/offline feature computation for real-time inference.; Optimizing inference latency and throughput using techniques like quantization, pruning, or batching..