F · Fine-tuning · first-contact resolution

Fine-tuning

Further training a base model on a small task-specific dataset.

In one sentence

Fine-tuning is the practice of taking a pre-trained foundation model and continuing its training on a smaller, task-specific dataset so it adopts a domain’s vocabulary, style or behaviour.

When it matters

When prompt engineering plus retrieval plateau and a domain has a recognisable register that the model keeps missing.

A real-world example

For a vending-machine fleet operator, fine-tuning on 6,000 anonymised past tickets gave a 12-point lift in first-contact resolution over RAG alone.

Curated by helpcode research team · Last reviewed 2026-05-22