RAG vs. Fine-Tuning
Both are ways to make an LLM work with a company's own information, and JDs often treat them as the same skill. RAG looks up relevant documents at the moment a question is asked and hands them to the model, so the model itself doesn't change. Fine-tuning retrains the model on examples, changing how it behaves. RAG is far more common. Fine-tuning needs more ML depth.
How to tell them apart on a resume
RAG
Vector databases, embeddings, chunking, search, LangChain or LlamaIndex, and 'chat with your documents'. Mostly software engineering skills.
Fine-Tuning
Training datasets, LoRA, GPUs, Hugging Face, evaluation before and after, and changing the model's tone, format or specialist skill.
The question that settles it
“Did you give the model documents to look things up in, or did you retrain the model itself on examples?”
Read the full definitions
Open the full tool for the other look-alike pairs, role profiles, and the JD decoder.