Generative AI & MLOps

What is an LLM?

An LLM (Large Language Model) is a large-scale language model trained on massive text corpora to predict the next token. This predictive capability enables it to generate coherent text, answer questions, summarize, translate, and reason. GPT-4, Claude, Gemini, and Llama are examples of LLMs.

What is RAG?

RAG (Retrieval-Augmented Generation) is an architecture that enriches an LLM's answers with documents retrieved in real time from a knowledge base (usually a vector store). This grounds responses in verifiable facts and bypasses the model's training cutoff date.

What is an AI agent?

An AI agent is a system where an LLM can plan actions, use external tools (web search, APIs, code, databases), and iterate to achieve a goal. Unlike a simple chatbot, an agent executes multi-step workflows autonomously.

What is MLOps?

MLOps (Machine Learning Operations) is the set of practices to industrialize the ML model lifecycle: development, testing, versioning, deployment, monitoring, and retraining. It applies DevOps principles to machine learning to take models from notebook to production.

What is the difference between MLOps and LLMOps?

MLOps applies to classical ML models (custom training, batch or real-time inference). LLMOps is specific to LLMs and GenAI: prompt management, output evaluation, fine-tuning, RAG, API cost management, and hallucination governance. Both share principles but differ in tooling and concerns.