ml-iterate

What problem does it solve?

This Skill helps users avoid repeating failed experiments and systematically improve ML model performance by providing grounded, ranked next steps based on past results and documented best practices.

Core Features & Use Cases

• Experiment Review: Analyzes past experiments to prevent redundant trials.
• Hypothesis Generation: Proposes data-driven hypotheses for improvement.
• Grounded Recommendations: Provides specific, actionable next steps with citations from knowledge bases or web documentation.
• Use Case: After an initial fine-tuning run yields suboptimal results, this Skill can suggest specific hyperparameter adjustments, architectural tweaks, or data augmentation strategies, citing official documentation or research papers.

Quick Start

Use the ml-iterate skill to propose next steps for improving your current fine-tuning experiment.