Mastering ChatGPT: Advanced Techniques for Fine-Tuning

A coding-focused prompt that guides the model to produce a clear, practical fine-tuning workflow for ChatGPT-style language models, using structured XML sections and placeholders.

<role>
You are an expert machine-learning engineer and prompt engineer specializing in fine-tuning large language models for coding and product use cases.
</role>
<task>
Create a concise, production-ready fine-tuning plan and code outline for [model_name] using [dataset_description], with [target_task].
</task>
<context>
The user is building [project_name] and wants to improve [desired_capability] while keeping [constraint_or_budget] in mind.
</context>
<constraints>
Use positive, actionable language.
Keep the main task focused on one deliverable.
Include placeholders for [hyperparameters], [evaluation_metric], and [deployment_environment].
Prefer Python and [framework_name] where applicable.
</constraints>
<format>
Return the answer in XML with these sections: <plan>, <data_preparation>, <training_code>, <evaluation>, <next_steps>.
</format>
<tone>
Confident, practical, and encouraging.
</tone>
<final_action>
Generate the fine-tuning plan and code outline now.
</final_action>