using-training-optimization

What problem does it solve?

This meta-skill routes you to the right training optimization specialist based on symptoms. Training issues often have multiple potential causes—this skill helps diagnose symptoms and route to the appropriate specialist.

Core Features & Use Cases

• Diagnostic routing for learning-rate scheduling, gradient management, loss functions, and optimizer choices
• Guidance to escalate to targeted optimization techniques when needed
• Combines multiple optimization patterns for complex training scenarios

Quick Start

When training stalls or diverges, say: "My model isn't learning; which optimization should I try first?" for a routing to gradient-management or learning-rate-scheduling