Can this skill help with both training and deployment simultaneously?

Yes, it handles cross-cutting scenarios by routing you to multiple specialized packs in a logical execution order, such as training-optimization followed by production-deployment.

When should I load the using-ai-engineering skill?

Load this skill whenever you are starting an AI/ML task but are unsure which specific domain-specialized pack (like RL, LLM, or PyTorch) is most appropriate for your current codebase.

How does this skill handle ambiguous prompts like 'model not working'?

It triggers a mandatory clarification process to determine if the issue lies in the architecture choice, training process, or deployment infrastructure before routing you to a specialized tool.

How does it handle urgent production issues?

It follows a specific emergency protocol that prioritizes fast, systematic diagnosis over guessing, ensuring that high-pressure situations are resolved using the correct technical foundations.

AI Engineering & ML Strategy Router

Name: AI Engineering & ML Strategy Router
Author: tachyon-beep

bytachyon-beep

•

データサイエンスとML

Routes AI and machine learning tasks to specialized Yzmir engineering packs based on specific project requirements and technical domains.

This skill acts as a central intelligence hub for AI/ML engineering within the Yzmir ecosystem. It prevents trial-and-error by systematically diagnosing your specific machine learning needs—whether it's foundational PyTorch issues, training instability, Reinforcement Learning (RL), LLM fine-tuning, or production deployment. By using a rigorous routing decision tree and mandatory clarification triggers, it ensures that Claude applies the most relevant domain expertise and best practices to your project while resisting common pitfalls like sunk-cost bias or authority pressure.

主な機能

01Systematic routing between specialized AI domains including LLMs, RL, and PyTorch

02Mandatory clarification triggers for ambiguous machine learning queries

03Domain-specific decision trees for training vs. deployment optimization

045 GitHub stars

05Multi-domain cross-cutting scenario mapping for complex workflows

06Pressure-resistant protocols for emergency and high-stakes engineering environments

ユースケース

01Starting a complex AI project and needing to identify the correct technical framework

02Debugging failing training loops or unstable neural network convergence

03Transitioning an AI model from experimental training to a production environment

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add tachyon-beep/skillpacks using-ai-engineering

For use in Claude.ai and ChatGPT

Download Skill