skills.rest
marcgreen
marcgreenCOMMUNITY
Explore all skills by this author →
GitHub

finetune-design

Create design artifacts for fine-tuning LLMs in multi-turn conversations.

4|1|Updated Dec 28, 2025
Search Tags:#prompt engineering#llm#fine-tuning#data generation#model selection#evaluation rubric
One-click install
npx skills add https://github.com/marcgreen/therapy-coach-finetune --skill finetune-design
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: finetune-design
Source: https://github.com/marcgreen/therapy-coach-finetune/tree/main/.claude/skills/finetune-design
Command: npx skills add https://github.com/marcgreen/therapy-coach-finetune --skill finetune-design

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill streamlines the complex process of preparing to fine-tune a Large Language Model (LLM) for multi-turn conversations, ensuring all necessary design artifacts are created before data generation begins.

Core Features & Use Cases

  • Systematic Design: Guides users through selecting models, defining token economics, creating taxonomies, designing evaluation rubrics, and crafting prompts.
  • Expert Validation: Integrates "Expert Role-Play Critique" to stress-test designs and identify blind spots.
  • Use Case: Before starting a project to fine-tune a model for customer support, use this Skill to define the types of customer issues, the desired quality standards, and the user personas to simulate, ensuring a robust foundation for data generation.

Quick Start

Use the finetune-design skill to begin planning your LLM fine-tuning project by designing the necessary artifacts.

Frequently Asked Questions about finetune-design

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I design an LLM fine-tuning project for multi-turn conversations?▼

To design an LLM fine-tuning project, systematically create essential artifacts covering model selection, token economics, input taxonomy, persona templates, and prompt design before generating training data.

What artifacts do I need before generating data to fine-tune an LLM?▼

Before generating data to fine-tune an LLM, you need artifacts including an input taxonomy, evaluation rubric, persona templates, prompt designs, and base model evaluation metrics to ensure a robust foundation.

How do I create an evaluation rubric for fine-tuning a customer support LLM?▼

Creating an evaluation rubric for fine-tuning a customer support LLM involves defining desired quality standards, categorizing customer issue types, and establishing metrics to evaluate base model performance.

Can I use expert role-play critique to validate my LLM fine-tuning design?▼

Yes, you can use expert role-play critique to validate your LLM fine-tuning design by stress-testing your artifacts against simulated user personas to identify blind spots and design flaws.

What is the best way to select a base model and define token economics for fine-tuning?▼

The best way to select a base model and define token economics for fine-tuning is to systematically evaluate your multi-turn conversation requirements against base model performance and projected token limits.

Related Skills

Explore More Software Engineering Skills→
HahyeonJeonHahyeonJeon

_agent-evaluation-project

Evaluate Gobbi agent definitions for purpose, scope, and routing alignment.

Community
Intermediate
pmcdowallpmcdowall

fine-tuning-with-trl

Align large language models via reinforcement learning and preference-based fine-tuning workflows.

Community
Advanced
JCE-Joshhh77JCE-Joshhh77
16

ai-optimization

Optimize AI/LLM usage through token efficiency, model selection, and prompt design.

Community
Advanced
fanguyunfanguyun

prompt-engineer

Develop, optimize, and evaluate LLM prompts using Python, Markdown, and YAML.

Community
Advanced