Glossary · 5 · Building with AI: APIs, tools and agents
Fine-tuning
Also known as: LoRA, Model customization
Fine-tuning is the further training of a pretrained model on a smaller, task-specific dataset so that it adopts a particular behavior, format, style or domain vocabulary more reliably than prompting alone achieves.
- Expert
- Technical project managers
- Developers
In one sentence
Fine-tuning explained: training a pretrained model on your examples for style or format — and why RAG usually comes first for knowledge.
Example
A company fine-tunes a small open-weight model on 5,000 approved short descriptions so that it writes new ones in exactly the house style.
Why it matters on your learning path
- Technical project managers: Fine-tuning needs curated training data, evaluation and retraining when the base model changes; plan it as an ongoing effort.
- Developers: Try prompting, few-shot examples and RAG first. Parameter-efficient methods such as LoRA make fine-tuning affordable.