Retrieval-Augmented Generation (RAG)
An AI architecture that retrieves verified documents before generating answers, which sharply reduces hallucinations.
Vector Embeddings
High-dimensional mathematical representations of text that capture semantic meaning and intent.
Cosine Similarity
A metric used to measure how semantically similar two text vectors are regardless of their length.
Grounded Answers
Architectural constraints that restrict an AI model to verified sources, so it is far less likely to invent false facts or unauthorized discounts.
Ticket Deflection
The percentage of incoming customer service inquiries resolved autonomously without human agent involvement.
WISMO (Where Is My Order?)
One of the most common e-commerce support inquiries, asking for shipment tracking and package status.
Document Chunking
The process of splitting large documents into smaller semantic text blocks for vector embedding and retrieval.
Prompt Injection
A security vulnerability where malicious user input attempts to override system instructions or extract confidential data.