Glossary1
Clear definitions of key applied AI terms. No academic jargon, no hype. For executives who need to make decisions, not specialists who already know the field.
RAG
A technique that enables a language model (LLM) to respond using your company’s documents, not just its general training. The system first searches a proprietary knowledge base…
AI agent
A system that observes an environment, decides on an action, and carries it out using digital tools. Unlike a chatbot (which only converses) or RPA (which only executes a fixed…
LLM
An artificial intelligence model trained on large volumes of text that generates natural language statistically. The best known are GPT (OpenAI), Claude (Anthropic), and Gemini…
Fine-tuning
The process of continuing to train a base model with examples specific to your domain so it can learn a particular style, format, or vocabulary. It requires a curated dataset…
Prompt Engineering
The discipline of designing instructions for a language model to get the desired response. It’s not just “writing nicely”: it involves structuring context, examples, output…
MLOps
The discipline of operationalizing machine learning models in production: model and data versioning, drift monitoring, automated retraining, rollback, observability, and…
Hallucination
A response generated by an AI model that sounds convincing but is factually incorrect or made up. It’s not a bug: it’s a natural consequence of how LLMs generate text (statistical…
Embedding
A numerical representation of text (or an image) as a vector with hundreds or thousands of dimensions, where the distance between vectors reflects semantic similarity. It’s the…